chore(ui): sync sentry config with master

.env

@@ -72,8 +72,8 @@ NEO4J_APOC_IMPORT_FILE_ENABLED=false
 NEO4J_APOC_IMPORT_FILE_USE_NEO4J_CONFIG=true
 NEO4J_APOC_TRIGGER_ENABLED=false
 NEO4J_DBMS_CONNECTOR_BOLT_LISTEN_ADDRESS=0.0.0.0:7687
-# Neo4j Prowler settings
-ATTACK_PATHS_BATCH_SIZE=1000
+# Attack Paths graph settings
+ATTACK_PATHS_GRAPH_MUTATION_BATCH_SIZE=1000
 ATTACK_PATHS_SERVICE_UNAVAILABLE_MAX_RETRIES=3
 ATTACK_PATHS_READ_QUERY_TIMEOUT_SECONDS=30
 ATTACK_PATHS_MAX_CUSTOM_QUERY_NODES=250
@@ -145,19 +145,20 @@ DJANGO_BROKER_VISIBILITY_TIMEOUT=86400
 DJANGO_SENTRY_DSN=
 DJANGO_THROTTLE_TOKEN_OBTAIN=50/minute
 
-# Sentry for the web app (server + browser). Empty/unset UI_SENTRY_DSN ⇒
-# Sentry disabled, zero egress. SENTRY_RELEASE (unprefixed) feeds the web app's
-# server/edge SDKs.
+# Sentry for the web app (server + browser). The UI_SENTRY_* values load only
+# when UI_SENTRY_ENABLED="true"; without it they are ignored (default off, zero
+# egress). The deprecated NEXT_PUBLIC_SENTRY_DSN still activates Sentry without
+# the flag. SENTRY_RELEASE (unprefixed) feeds the web app's server/edge SDKs.
 UI_SENTRY_DSN=
 UI_SENTRY_ENVIRONMENT=local
 SENTRY_RELEASE=local
 # Reserved runtime public config (registered now; no UI consumer yet)
-# POSTHOG_KEY=
-# POSTHOG_HOST=
+# UI_POSTHOG_KEY=
+# UI_POSTHOG_HOST=
 # REO_DEV_CLIENT_ID=
 
 #### Prowler release version ####
-NEXT_PUBLIC_PROWLER_RELEASE_VERSION=v5.31.0
+NEXT_PUBLIC_PROWLER_RELEASE_VERSION=v5.35.0
 
 # Social login credentials
 SOCIAL_GOOGLE_OAUTH_CALLBACK_URL="${AUTH_URL}/api/auth/callback/google"

.github/actions/setup-python-uv/action.yml

@@ -46,7 +46,7 @@ runs:
       env:
         GITHUB_TOKEN: ${{ github.token }}
       run: |
-        LATEST_COMMIT=$(curl -sf \
+        LATEST_COMMIT=$(curl -sf --retry 3 --retry-all-errors --retry-delay 2 --retry-max-time 60 \
           -H "Authorization: Bearer ${GITHUB_TOKEN}" \
           -H "Accept: application/vnd.github+json" \
           "https://api.github.com/repos/prowler-cloud/prowler/commits/master" \
@@ -66,7 +66,7 @@ runs:
       env:
         GITHUB_TOKEN: ${{ github.token }}
       run: |
-        LATEST_COMMIT=$(curl -sf \
+        LATEST_COMMIT=$(curl -sf --retry 3 --retry-all-errors --retry-delay 2 --retry-max-time 60 \
           -H "Authorization: Bearer ${GITHUB_TOKEN}" \
           -H "Accept: application/vnd.github+json" \
           "https://api.github.com/repos/prowler-cloud/prowler/commits/master" \

.github/actions/trivy-scan/action.yml

@@ -63,7 +63,7 @@ runs:
         exit-code: '0'
         scanners: 'vuln'
         timeout: '5m'
-        version: 'v0.71.0'
+        version: 'v0.71.2'
 
     - name: Run Trivy vulnerability scan (SARIF)
       if: inputs.upload-sarif == 'true' && github.event_name == 'push'
@@ -76,7 +76,7 @@ runs:
         exit-code: '0'
         scanners: 'vuln'
         timeout: '5m'
-        version: 'v0.71.0'
+        version: 'v0.71.2'
 
     - name: Upload Trivy results to GitHub Security tab
       if: inputs.upload-sarif == 'true' && github.event_name == 'push'

.github/aw/actions-lock.json

@@ -5,10 +5,20 @@
       "version": "v8",
       "sha": "ed597411d8f924073f98dfc5c65a23a2325f34cd"
     },
+    "github/gh-aw-actions/setup@v0.81.6": {
+      "repo": "github/gh-aw-actions/setup",
+      "version": "v0.81.6",
+      "sha": "ba6380cc6e5be5d21677bebe04d52fb48e3abec7"
+    },
     "github/gh-aw/actions/setup@v0.43.23": {
       "repo": "github/gh-aw/actions/setup",
       "version": "v0.43.23",
       "sha": "9382be3ca9ac18917e111a99d4e6bbff58d0dccc"
+    },
+    "step-security/harden-runner@v2.20.0": {
+      "repo": "step-security/harden-runner",
+      "version": "v2.20.0",
+      "sha": "bf7454d06d71f1098171f2acdf0cd4708d7b5920"
     }
   }
 }

.github/dependabot.yml

@@ -1,138 +0,0 @@
-# To get started with Dependabot version updates, you'll need to specify which
-# package ecosystems to update and where the package manifests are located.
-# Please see the documentation for all configuration options:
-# https://docs.github.com/github/administering-a-repository/configuration-options-for-dependency-updates
-
-version: 2
-updates:
-  # v5
-  # - package-ecosystem: "pip"
-  #   directory: "/"
-  #   schedule:
-  #     interval: "monthly"
-  #   open-pull-requests-limit: 25
-  #   target-branch: master
-  #   labels:
-  #     - "dependencies"
-  #     - "pip"
-  #   cooldown:
-  #     default-days: 7
-
-  # Dependabot Updates are temporary disabled - 2025/03/19
-  # - package-ecosystem: "pip"
-  #   directory: "/api"
-  #   schedule:
-  #     interval: "daily"
-  #   open-pull-requests-limit: 10
-  #   target-branch: master
-  #   labels:
-  #     - "dependencies"
-  #     - "pip"
-  #     - "component/api"
-
-  - package-ecosystem: "github-actions"
-    directory: "/"
-    schedule:
-      interval: "monthly"
-    open-pull-requests-limit: 25
-    target-branch: master
-    labels:
-      - "dependencies"
-      - "github_actions"
-    cooldown:
-      default-days: 7
-
-  # Dependabot Updates are temporary disabled - 2025/03/19
-  # - package-ecosystem: "npm"
-  #   directory: "/ui"
-  #   schedule:
-  #     interval: "daily"
-  #   open-pull-requests-limit: 10
-  #   target-branch: master
-  #   labels:
-  #     - "dependencies"
-  #     - "npm"
-  #     - "component/ui"
-
-  - package-ecosystem: "docker"
-    directory: "/"
-    schedule:
-      interval: "monthly"
-    open-pull-requests-limit: 25
-    target-branch: master
-    labels:
-      - "dependencies"
-      - "docker"
-    cooldown:
-      default-days: 7
-
-  # - package-ecosystem: "pre-commit"
-  #   directory: "/"
-  #   schedule:
-  #     interval: "monthly"
-  #   open-pull-requests-limit: 25
-  #   target-branch: master
-  #   labels:
-  #     - "dependencies"
-  #     - "pre-commit"
-  #   cooldown:
-  #     default-days: 7
-
-  # Dependabot Updates are temporary disabled - 2025/04/15
-  # v4.6
-  # - package-ecosystem: "pip"
-  #   directory: "/"
-  #   schedule:
-  #     interval: "weekly"
-  #   open-pull-requests-limit: 10
-  #   target-branch: v4.6
-  #   labels:
-  #     - "dependencies"
-  #     - "pip"
-  #     - "v4"
-
-  # - package-ecosystem: "github-actions"
-  #   directory: "/"
-  #   schedule:
-  #     interval: "weekly"
-  #   open-pull-requests-limit: 10
-  #   target-branch: v4.6
-  #   labels:
-  #     - "dependencies"
-  #     - "github_actions"
-  #     - "v4"
-
-  # - package-ecosystem: "docker"
-  #   directory: "/"
-  #   schedule:
-  #     interval: "weekly"
-  #   open-pull-requests-limit: 10
-  #   target-branch: v4.6
-  #   labels:
-  #     - "dependencies"
-  #     - "docker"
-  #     - "v4"
-
-  # Dependabot Updates are temporary disabled - 2025/03/19
-  # v3
-  # - package-ecosystem: "pip"
-  #   directory: "/"
-  #   schedule:
-  #     interval: "monthly"
-  #   open-pull-requests-limit: 10
-  #   target-branch: v3
-  #   labels:
-  #     - "dependencies"
-  #     - "pip"
-  #     - "v3"
-
-  # - package-ecosystem: "github-actions"
-  #   directory: "/"
-  #   schedule:
-  #     interval: "monthly"
-  #   open-pull-requests-limit: 10
-  #   target-branch: v3
-  #   labels:
-  #     - "dependencies"
-  #     - "github_actions"
-  #     - "v3"

.github/pull_request_template.md

@@ -18,8 +18,9 @@ Please add a detailed description of how to review this PR.
 
 <summary><b>Community Checklist</b></summary>
 
-- [ ] This feature/issue is listed in [here](https://github.com/prowler-cloud/prowler/issues?q=sort%3Aupdated-desc+is%3Aissue+is%3Aopen) or roadmap.prowler.com
-- [ ] Is it assigned to me, if not, request it via the issue/feature in [here](https://github.com/prowler-cloud/prowler/issues?q=sort%3Aupdated-desc+is%3Aissue+is%3Aopen) or [Prowler Community Slack](goto.prowler.com/slack)
+- [ ] This feature/issue is listed in the [open issues](https://github.com/prowler-cloud/prowler/issues?q=sort%3Aupdated-desc+is%3Aissue+is%3Aopen) or roadmap.prowler.com
+- [ ] Is it assigned to me, if not, request it via the [open issues](https://github.com/prowler-cloud/prowler/issues?q=sort%3Aupdated-desc+is%3Aissue+is%3Aopen) or [Prowler Community Slack](https://goto.prowler.com/slack)
+- [ ] I have reviewed the [open pull requests](https://github.com/prowler-cloud/prowler/pulls?q=sort%3Aupdated-desc+is%3Apr+is%3Aopen) and confirmed there is no existing PR that implements the same outcome
 
 </details>
 
@@ -27,8 +28,8 @@ Please add a detailed description of how to review this PR.
 - [ ] Review if the code is being covered by tests.
 - [ ] Review if code is being documented following this specification https://github.com/google/styleguide/blob/gh-pages/pyguide.md#38-comments-and-docstrings
 - [ ] Review if backport is needed.
-- [ ] Review if is needed to change the [Readme.md](https://github.com/prowler-cloud/prowler/blob/master/README.md)
-- [ ] Ensure new entries are added to [CHANGELOG.md](https://github.com/prowler-cloud/prowler/blob/master/prowler/CHANGELOG.md), if applicable.
+- [ ] Review if is needed to change the [README.md](https://github.com/prowler-cloud/prowler/blob/master/README.md)
+- [ ] Ensure a changelog fragment is added under [prowler/changelog.d/](https://github.com/prowler-cloud/prowler/tree/master/prowler/changelog.d), if applicable.
 
 #### SDK/CLI
 - Are there new checks included in this PR? Yes / No
@@ -40,7 +41,7 @@ Please add a detailed description of how to review this PR.
 - [ ] Screenshots/Video of the functionality flow (if applicable) - Mobile (X < 640px)
 - [ ] Screenshots/Video of the functionality flow (if applicable) - Table (640px > X < 1024px)
 - [ ] Screenshots/Video of the functionality flow (if applicable) - Desktop (X > 1024px)
-- [ ] Ensure new entries are added to [CHANGELOG.md](https://github.com/prowler-cloud/prowler/blob/master/ui/CHANGELOG.md), if applicable.
+- [ ] Ensure a changelog fragment is added under [ui/changelog.d/](https://github.com/prowler-cloud/prowler/tree/master/ui/changelog.d), if applicable.
 
 #### API
 - [ ] All issue/task requirements work as expected on the API
@@ -50,7 +51,11 @@ Please add a detailed description of how to review this PR.
 - [ ] Any other relevant evidence of the implementation (if applicable)
 - [ ] Verify if API specs need to be regenerated.
 - [ ] Check if version updates are required (e.g., specs, uv, etc.).
-- [ ] Ensure new entries are added to [CHANGELOG.md](https://github.com/prowler-cloud/prowler/blob/master/api/CHANGELOG.md), if applicable.
+- [ ] Ensure a changelog fragment is added under [api/changelog.d/](https://github.com/prowler-cloud/prowler/tree/master/api/changelog.d), if applicable.
+
+#### MCP Server
+- [ ] All issue/task requirements work as expected on the MCP Server
+- [ ] Ensure a changelog fragment is added under [mcp_server/changelog.d/](https://github.com/prowler-cloud/prowler/tree/master/mcp_server/changelog.d), if applicable.
 
 ### License
 

.github/renovate.json

@@ -23,6 +23,10 @@
   "prConcurrentLimit": 20,
   "prHourlyLimit": 10,
   "vulnerabilityAlerts": {
+    "labels": [
+      "dependencies",
+      "security"
+    ],
     "prHourlyLimit": 0,
     "prConcurrentLimit": 0
   },
@@ -38,7 +42,7 @@
       "schedule": [
         "* 22-23,0-5 1 * *"
       ],
-      "enabled": false
+      "enabled": true
     },
     {
       "description": "Minors: 8th of every 3 months, Madrid overnight window (22:00-06:00)",
@@ -48,7 +52,7 @@
       "schedule": [
         "* 22-23,0-5 8 */3 *"
       ],
-      "enabled": false
+      "enabled": true
     },
     {
       "description": "Majors: 15th of every 3 months, Madrid overnight window",
@@ -58,6 +62,13 @@
       "schedule": [
         "* 22-23,0-5 15 */3 *"
       ],
+      "enabled": true
+    },
+    {
+      "description": "gh-aw compiled lock files - generated by 'gh aw compile', action pins must match the compiler version, never bump directly",
+      "matchFileNames": [
+        ".github/workflows/*.lock.yml"
+      ],
       "enabled": false
     },
     {

.github/scripts/changelog_attribution.py

@@ -0,0 +1,154 @@
+#!/usr/bin/env python3
+"""Rename changelog fragments to their PR number before running towncrier.
+
+For every <slug>.<type>.md in <component_dir>/changelog.d/, find the commit that
+added it, resolve its PR via the GitHub API (falling back to the squash-commit
+subject), and `git mv` it to <PR>.<type>.md so towncrier renders the PR link.
+Unresolvable fragments become +<slug>.<type>.md orphans (rendered without link).
+"""
+
+import argparse
+import json
+import os
+import re
+import subprocess
+import sys
+import urllib.error
+import urllib.request
+
+FRAGMENT_RE = re.compile(
+    r"^(?P<slug>[A-Za-z0-9][A-Za-z0-9._-]*?)"
+    r"\.(?P<type>added|changed|deprecated|removed|fixed|security)"
+    r"(?:\.(?P<counter>[0-9]+))?\.md$"
+)
+SUBJECT_PR_RE = re.compile(r" \(#([0-9]+)\)$")
+IGNORED_FILES = {".gitkeep", "README.md"}
+API_TIMEOUT_SECONDS = 10
+
+
+def git(*args: str) -> str:
+    result = subprocess.run(["git", *args], check=True, capture_output=True, text=True)
+    return result.stdout.strip()
+
+
+def find_adding_commit(path: str) -> str | None:
+    """Find the commit that added a file, following renames.
+
+    Falls back to a plain (no --follow) lookup: rename detection can lose the
+    add event for degenerate content (e.g. files identical to many others).
+    """
+    sha = git("log", "--follow", "--diff-filter=A", "--format=%H", "-1", "--", path)
+    if not sha:
+        sha = git("log", "--diff-filter=A", "--format=%H", "-1", "--", path)
+    return sha or None
+
+
+def pr_from_api(repo: str, sha: str) -> int | None:
+    """Resolve the PR associated with a commit via the GitHub API.
+
+    Returns None on any network/API failure so the caller can fall back to
+    parsing the squash-commit subject.
+    """
+    url = f"https://api.github.com/repos/{repo}/commits/{sha}/pulls"
+    headers = {
+        "Accept": "application/vnd.github+json",
+        "X-GitHub-Api-Version": "2022-11-28",
+        "User-Agent": "prowler-changelog-attribution",
+    }
+    token = os.environ.get("GITHUB_TOKEN")
+    if token:
+        headers["Authorization"] = f"Bearer {token}"
+    request = urllib.request.Request(url, headers=headers)
+    try:
+        with urllib.request.urlopen(request, timeout=API_TIMEOUT_SECONDS) as response:
+            pulls = json.load(response)
+    except (urllib.error.URLError, TimeoutError, json.JSONDecodeError):
+        return None
+    if isinstance(pulls, list) and pulls:
+        return pulls[0].get("number")
+    return None
+
+
+def pr_from_subject(sha: str) -> int | None:
+    subject = git("log", "-1", "--format=%s", sha)
+    match = SUBJECT_PR_RE.search(subject)
+    return int(match.group(1)) if match else None
+
+
+def unique_destination(directory: str, base_name: str, fragment_type: str) -> str:
+    """Return a non-colliding fragment path, appending a numeric counter if needed."""
+    candidate = os.path.join(directory, f"{base_name}.{fragment_type}.md")
+    counter = 0
+    while os.path.exists(candidate):
+        counter += 1
+        candidate = os.path.join(directory, f"{base_name}.{fragment_type}.{counter}.md")
+    return candidate
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("component_dir", help="Component directory, e.g. prowler")
+    parser.add_argument("--repo", default="prowler-cloud/prowler")
+    parser.add_argument(
+        "--no-api",
+        action="store_true",
+        help="Skip the GitHub API and resolve PRs from commit subjects only",
+    )
+    args = parser.parse_args()
+
+    fragments_dir = os.path.join(args.component_dir, "changelog.d")
+    if not os.path.isdir(fragments_dir):
+        print(f"::error::Fragments directory not found: {fragments_dir}")
+        return 1
+
+    malformed = []
+    to_process = []
+    for name in sorted(os.listdir(fragments_dir)):
+        if name in IGNORED_FILES or name.startswith("+"):
+            continue
+        match = FRAGMENT_RE.match(name)
+        if not match:
+            malformed.append(name)
+            continue
+        if match.group("slug").isdigit():
+            continue
+        to_process.append((name, match))
+
+    if malformed:
+        for name in malformed:
+            print(
+                f"::error::Malformed fragment filename in {fragments_dir}: {name} "
+                "(expected <slug>.<type>.md with type one of added|changed|"
+                "deprecated|removed|fixed|security)"
+            )
+        return 1
+
+    for name, match in to_process:
+        slug, fragment_type = match.group("slug"), match.group("type")
+
+        path = os.path.join(fragments_dir, name)
+        sha = find_adding_commit(path)
+        pr_number = None
+        if sha:
+            if not args.no_api:
+                pr_number = pr_from_api(args.repo, sha)
+            if pr_number is None:
+                pr_number = pr_from_subject(sha)
+
+        if pr_number is not None:
+            destination = unique_destination(
+                fragments_dir, str(pr_number), fragment_type
+            )
+        else:
+            destination = unique_destination(fragments_dir, f"+{slug}", fragment_type)
+            print(
+                f"::warning::Could not resolve a PR for {path}; renamed to "
+                f"{os.path.basename(destination)} (entry will render without a PR link)"
+            )
+        git("mv", path, destination)
+        print(f"{path} -> {destination}")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

.github/towncrier/template.md.jinja

@@ -0,0 +1,13 @@
+{% set category_order = definitions.keys() %}
+{% for section, _ in sections.items() %}
+{% for category in category_order if category in sections[section] %}
+### {{ definitions[category]['name'] }}
+
+{% for text, values in sections[section][category].items() -%}
+- {{ text }}{% if values %} {{ values|join(', ') }}{% endif %}{{ "\n" }}
+{%- endfor %}
+
+{% endfor %}
+{% endfor %}
+---
+{{ "\n" }}

.github/workflows/api-code-quality.yml

@@ -35,7 +35,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -62,6 +62,7 @@ jobs:
             api/docs/**
             api/README.md
             api/CHANGELOG.md
+            api/changelog.d/**
             api/AGENTS.md
 
       - name: Setup Python with uv

.github/workflows/api-codeql.yml

@@ -9,6 +9,7 @@ on:
       - 'api/**'
       - '.github/workflows/api-codeql.yml'
       - '.github/codeql/api-codeql-config.yml'
+      - '!api/CHANGELOG.md'
   pull_request:
     branches:
       - 'master'
@@ -17,6 +18,7 @@ on:
       - 'api/**'
       - '.github/workflows/api-codeql.yml'
       - '.github/codeql/api-codeql-config.yml'
+      - '!api/CHANGELOG.md'
   schedule:
     - cron: '00 12 * * *'
 
@@ -44,7 +46,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/api-container-build-push.yml

@@ -46,7 +46,7 @@ jobs:
       contents: read
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
 
@@ -65,7 +65,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -108,7 +108,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -175,7 +175,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -215,7 +215,7 @@ jobs:
 
       - name: Install regctl
         if: always()
-        uses: regclient/actions/regctl-installer@da9319db8e44e8b062b3a147e1dfb2f574d41a03 # main
+        uses: regclient/actions/regctl-installer@9a2d4216180dbb3e2dccfa60d2dd4afd98e42ec5 # main
 
       - name: Cleanup intermediate architecture tags
         if: always()
@@ -236,7 +236,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -272,27 +272,3 @@ jobs:
           payload-file-path: "./.github/scripts/slack-messages/container-release-completed.json"
           step-outcome: ${{ steps.outcome.outputs.outcome }}
           update-ts: ${{ needs.notify-release-started.outputs.message-ts }}
-
-  trigger-deployment:
-    needs: [setup, container-build-push]
-    if: always() && github.event_name == 'push' && needs.setup.result == 'success' && needs.container-build-push.result == 'success'
-    runs-on: ubuntu-latest
-    timeout-minutes: 5
-    permissions:
-      contents: read
-
-    steps:
-      - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
-        with:
-          egress-policy: block
-          allowed-endpoints: >
-            api.github.com:443
-
-      - name: Trigger API deployment
-        uses: peter-evans/repository-dispatch@28959ce8df70de7be546dd1250a005dd32156697 # v4.0.1
-        with:
-          token: ${{ secrets.PROWLER_BOT_ACCESS_TOKEN }}
-          repository: ${{ secrets.CLOUD_DISPATCH }}
-          event-type: api-prowler-deployment
-          client-payload: '{"sha": "${{ github.sha }}", "short_sha": "${{ needs.setup.outputs.short-sha }}"}'

.github/workflows/api-container-checks.yml

@@ -33,7 +33,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -69,7 +69,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -92,6 +92,8 @@ jobs:
             _http._tcp.deb.debian.org:443
             powershellinfraartifacts-gkhedzdeaghdezhr.z01.azurefd.net:443
             get.trivy.dev:443
+            raw.githubusercontent.com:443
+            releases.astral.sh:443
 
       - name: Checkout repository
         uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
@@ -108,6 +110,7 @@ jobs:
             api/docs/**
             api/README.md
             api/CHANGELOG.md
+            api/changelog.d/**
             api/AGENTS.md
 
       - name: Set up Docker Buildx

.github/workflows/api-security.yml

@@ -43,7 +43,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -77,6 +77,7 @@ jobs:
             api/docs/**
             api/README.md
             api/CHANGELOG.md
+            api/changelog.d/**
             api/AGENTS.md
 
       - name: Setup Python with uv

.github/workflows/api-tests.yml

@@ -48,7 +48,7 @@ jobs:
 
     services:
       postgres:
-        image: postgres:17@sha256:2cd82735a36356842d5eb1ef80db3ae8f1154172f0f653db48fde079b2a0b7f7
+        image: postgres:17@sha256:5c855ad7b85e68e48a62f34662853f38b57c1c1d80f3a927ab58034fd6d31c5e
         env:
           POSTGRES_HOST: ${{ env.POSTGRES_HOST }}
           POSTGRES_PORT: ${{ env.POSTGRES_PORT }}
@@ -63,7 +63,7 @@ jobs:
           --health-timeout 5s
           --health-retries 5
       valkey:
-        image: valkey/valkey:7-alpine3.19
+        image: valkey/valkey:7-alpine3.19@sha256:4054fe7fc607b9326ac7c4691ed26e9670d2ff17a9fb28c2577adecf928acbcc
         env:
           VALKEY_HOST: ${{ env.VALKEY_HOST }}
           VALKEY_PORT: ${{ env.VALKEY_PORT }}
@@ -78,7 +78,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -111,6 +111,7 @@ jobs:
             api/docs/**
             api/README.md
             api/CHANGELOG.md
+            api/changelog.d/**
             api/AGENTS.md
 
       - name: Setup Python with uv

.github/workflows/backport.yml

@@ -30,7 +30,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/bump-version.yml

@@ -29,7 +29,7 @@ jobs:
       patch_version: ${{ steps.detect.outputs.patch_version }}
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -75,7 +75,7 @@ jobs:
       pull-requests: write
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -202,7 +202,7 @@ jobs:
       pull-requests: write
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -307,7 +307,7 @@ jobs:
       pull-requests: write
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/check-test-init-files.yml

@@ -0,0 +1,35 @@
+name: 'Tools: Check Test Init Files'
+
+on:
+  pull_request:
+    branches:
+      - 'master'
+      - 'v5.*'
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+permissions: {}
+
+jobs:
+  check-test-init-files:
+    if: github.repository == 'prowler-cloud/prowler'
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    permissions:
+      contents: read
+
+    steps:
+      - name: Harden the runner (Audit all outbound calls)
+        uses: step-security/harden-runner@fa2e9d605c4eeb9fcad4c99c224cee0c6c7f3594 # v2.16.0
+        with:
+          egress-policy: audit
+
+      - name: Checkout repository
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        with:
+          persist-credentials: false
+
+      - name: Check for __init__.py files in test directories
+        run: python3 scripts/check_test_init_files.py .

.github/workflows/ci-zizmor.yml

@@ -36,7 +36,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/comment-label-update.yml

@@ -22,7 +22,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/compile-changelogs.yml

@@ -0,0 +1,527 @@
+name: 'Tools: Compile Changelogs'
+
+run-name: 'Compile changelogs for Prowler ${{ inputs.prowler_version }}'
+
+on:
+  workflow_dispatch:
+    inputs:
+      prowler_version:
+        description: 'Prowler version being released (e.g., 5.31.0)'
+        required: true
+        type: string
+      target_branch:
+        description: 'Branch to compile on (master for minor releases, v5.X for patches)'
+        required: true
+        type: string
+      sdk_version:
+        description: 'SDK version override (empty = mirrors prowler_version; "skip" = hold this component back)'
+        required: false
+        type: string
+      api_version:
+        description: 'API version override (empty = auto-derive 1.<prowler_minor + 1>.<prowler_patch>; "skip" = hold back)'
+        required: false
+        type: string
+      ui_version:
+        description: 'UI version override (empty = auto-derive 1.<prowler_minor>.<prowler_patch>; "skip" = hold back)'
+        required: false
+        type: string
+      mcp_version:
+        description: 'MCP Server version override (empty = auto-derive from pending fragment types; "skip" = hold back)'
+        required: false
+        type: string
+
+concurrency:
+  group: ${{ github.workflow }}-${{ inputs.prowler_version }}
+  cancel-in-progress: false
+
+env:
+  PROWLER_VERSION: ${{ inputs.prowler_version }}
+  TARGET_BRANCH: ${{ inputs.target_branch }}
+  SDK_VERSION: ${{ inputs.sdk_version }}
+  API_VERSION: ${{ inputs.api_version }}
+  UI_VERSION: ${{ inputs.ui_version }}
+  MCP_VERSION: ${{ inputs.mcp_version }}
+
+permissions: {}
+
+jobs:
+  compile-changelogs:
+    if: github.event_name == 'workflow_dispatch' && github.repository == 'prowler-cloud/prowler'
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    permissions:
+      contents: write
+      pull-requests: write
+    steps:
+      - name: Harden the runner (Block outbound calls)
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
+        with:
+          egress-policy: block
+          allowed-endpoints: >
+            api.github.com:443
+            github.com:443
+            objects.githubusercontent.com:443
+            pypi.org:443
+            files.pythonhosted.org:443
+
+      - name: Checkout repository
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        with:
+          ref: ${{ inputs.target_branch }}
+          fetch-depth: 0 # PR attribution resolves each fragment's adding commit from history
+          token: ${{ secrets.PROWLER_BOT_ACCESS_TOKEN }}
+          persist-credentials: false
+
+      - name: Set up Python
+        uses: actions/setup-python@e797f83bcb11b83ae66e0230d6156d7c80228e7c # v6.0.0
+        with:
+          python-version: '3.12'
+
+      - name: Install towncrier
+        run: pip install --no-cache-dir towncrier==25.8.0
+
+      - name: Configure Git
+        run: |
+          git config --global user.name 'prowler-bot'
+          git config --global user.email '179230569+prowler-bot@users.noreply.github.com'
+
+      - name: Validate version inputs
+        run: |
+          if [[ ! "$PROWLER_VERSION" =~ ^[0-9]+\.[0-9]+\.[0-9]+$ ]]; then
+            echo "::error::Invalid prowler_version syntax: '$PROWLER_VERSION' (must be N.N.N)"
+            exit 1
+          fi
+          if [ "$TARGET_BRANCH" != "master" ] && [[ ! "$TARGET_BRANCH" =~ ^v[0-9]+\.[0-9]+$ ]]; then
+            echo "::error::Invalid target_branch syntax: '$TARGET_BRANCH' (must be 'master' or vN.N, e.g. v5.31)"
+            exit 1
+          fi
+          IFS=. read -r prowler_major prowler_minor prowler_patch <<< "$PROWLER_VERSION"
+          prowler_major=$((10#$prowler_major))
+          prowler_minor=$((10#$prowler_minor))
+          prowler_patch=$((10#$prowler_patch))
+          if [ "$prowler_patch" -eq 0 ]; then
+            if [ "$TARGET_BRANCH" != "master" ]; then
+              echo "::error::target_branch must be 'master' for Prowler ${PROWLER_VERSION}; got '${TARGET_BRANCH}'"
+              exit 1
+            fi
+          else
+            expected_target_branch="v${prowler_major}.${prowler_minor}"
+            if [ "$TARGET_BRANCH" != "$expected_target_branch" ]; then
+              echo "::error::target_branch must be '${expected_target_branch}' for Prowler ${PROWLER_VERSION}; got '${TARGET_BRANCH}'"
+              exit 1
+            fi
+          fi
+          for pair in "sdk_version:$SDK_VERSION" "api_version:$API_VERSION" "ui_version:$UI_VERSION" "mcp_version:$MCP_VERSION"; do
+            input_name="${pair%%:*}"
+            input_value="${pair#*:}"
+            if [ -n "$input_value" ] && [ "$input_value" != "skip" ] && [[ ! "$input_value" =~ ^[0-9]+\.[0-9]+\.[0-9]+$ ]]; then
+              echo "::error::Invalid $input_name syntax: '$input_value' (must be N.N.N, empty for auto-derivation, or 'skip')"
+              exit 1
+            fi
+          done
+
+      - name: Compile changelogs
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          set -euo pipefail
+
+          component_input() {
+            case "$1" in
+              prowler) echo "$SDK_VERSION" ;;
+              api) echo "$API_VERSION" ;;
+              ui) echo "$UI_VERSION" ;;
+              mcp_server) echo "$MCP_VERSION" ;;
+            esac
+          }
+
+          version_key() {
+            local version="$1"
+            local major minor patch
+            IFS=. read -r major minor patch <<< "$version"
+            printf '%06d.%06d.%06d' "$((10#$major))" "$((10#$minor))" "$((10#$patch))"
+          }
+
+          pending_fragments() {
+            find "$1/changelog.d" -maxdepth 1 -type f ! -name '.gitkeep' ! -name 'README.md' | sort
+          }
+
+          # The component's last released version is the first stamped heading
+          # of its CHANGELOG.md, the same source prepare-release.yml greps.
+          latest_released_version() {
+            grep -m1 -E '^## \[v?[0-9]+\.[0-9]+\.[0-9]+\]' "$1/CHANGELOG.md" | sed -E 's/^## \[v?([0-9]+\.[0-9]+\.[0-9]+)\].*/\1/'
+          }
+
+          has_removed_fragments() {
+            echo "$1" | grep -qE '\.removed(\.[0-9]+)?\.md$'
+          }
+
+          # Resolve every component's effective version before compiling
+          # anything, so a wrong input cannot leave the tree half-compiled.
+          # Empty input = auto-derive (latest released version + semver bump
+          # from the pending fragment types). 'skip' = hold the component back.
+          errors=0
+          compiling=""
+          for component in prowler api ui mcp_server; do
+            input=$(component_input "$component")
+            fragments=$(pending_fragments "$component")
+
+            if [ "$input" = "skip" ]; then
+              if [ -n "$fragments" ]; then
+                echo "::warning::${component}: held back by request; these pending fragments stay for a future release:"
+                echo "$fragments"
+              fi
+              continue
+            fi
+            if [ -n "$input" ] && [ -z "$fragments" ]; then
+              echo "::error::${component}: version input '$input' provided but ${component}/changelog.d/ has no pending fragments (wrong input?)"
+              errors=1
+              continue
+            fi
+            if [ -z "$fragments" ]; then
+              continue
+            fi
+
+            removed_fragments=false
+            if has_removed_fragments "$fragments"; then
+              removed_fragments=true
+            fi
+            current=$(latest_released_version "$component")
+            if [ -z "$current" ]; then
+              echo "::error::${component}: could not read the latest released version from ${component}/CHANGELOG.md; restore the released heading before compiling"
+              errors=1
+              continue
+            fi
+
+            if [ -n "$input" ]; then
+              effective="$input"
+              mode="explicit"
+              current_key=$(version_key "$current")
+              effective_key=$(version_key "$effective")
+              if [[ "$effective_key" < "$current_key" || "$effective_key" == "$current_key" ]]; then
+                echo "::error::${component}: explicit version '${effective}' must be greater than the latest released version (${current})"
+                errors=1
+                continue
+              fi
+            else
+              if [ "$removed_fragments" = "true" ]; then
+                echo "::error::${component}: pending 'removed' fragments imply a major bump (breaking change); provide its version input explicitly"
+                errors=1
+                continue
+              fi
+              # SDK, UI, and API versions are deterministic mirrors of the
+              # Prowler version (the scheme bump-version.yml codifies): the SDK
+              # mirrors it directly, the UI tracks 1.<minor>.<patch>, and the
+              # API is the independent 1.<minor + 1>.<patch> stream. Only the
+              # MCP Server has its own cadence, derived from fragment types.
+              IFS=. read -r _ prowler_minor prowler_patch <<< "$PROWLER_VERSION"
+              prowler_minor=$((10#$prowler_minor))
+              prowler_patch=$((10#$prowler_patch))
+              case "$component" in
+                prowler) effective="$PROWLER_VERSION" ;;
+                ui) effective="1.${prowler_minor}.${prowler_patch}" ;;
+                api) effective="1.$((prowler_minor + 1)).${prowler_patch}" ;;
+                mcp_server)
+                  IFS=. read -r major minor patch <<< "$current"
+                  major=$((10#$major))
+                  minor=$((10#$minor))
+                  patch=$((10#$patch))
+                  # Prowler patch releases (vN.N target) are maintenance
+                  # releases, so the MCP Server bumps patch regardless of
+                  # fragment types; a deliberate exception needs the explicit
+                  # version input.
+                  if [ "$TARGET_BRANCH" != "master" ]; then
+                    effective="${major}.${minor}.$((patch + 1))"
+                    if echo "$fragments" | grep -qE '\.(added|deprecated)(\.[0-9]+)?\.md$'; then
+                      echo "::warning::${component}: 'added'/'deprecated' fragments are shipping in a Prowler patch; auto-derived a patch bump (${current} -> ${effective}), pass the version input to override"
+                    fi
+                  elif echo "$fragments" | grep -qE '\.(added|changed|deprecated)(\.[0-9]+)?\.md$'; then
+                    effective="${major}.$((minor + 1)).0"
+                  else
+                    effective="${major}.${minor}.$((patch + 1))"
+                  fi
+                  ;;
+              esac
+              current_key=$(version_key "$current")
+              effective_key=$(version_key "$effective")
+              if [[ "$effective_key" < "$current_key" || "$effective_key" == "$current_key" ]]; then
+                echo "::error::${component}: auto-derived version '${effective}' is not greater than the latest released version (${current}); check prowler_version or pass the version input explicitly"
+                errors=1
+                continue
+              fi
+              mode="auto"
+              echo "::notice::${component}: version auto-derived ${current} -> ${effective}"
+            fi
+
+            if [ "$removed_fragments" = "true" ]; then
+              IFS=. read -r current_major _ <<< "$current"
+              current_major=$((10#$current_major))
+              IFS=. read -r effective_major effective_minor effective_patch <<< "$effective"
+              effective_major=$((10#$effective_major))
+              effective_minor=$((10#$effective_minor))
+              effective_patch=$((10#$effective_patch))
+              if [ "$effective_major" -le "$current_major" ] || [ "$effective_minor" -ne 0 ] || [ "$effective_patch" -ne 0 ]; then
+                echo "::error::${component}: removed fragments require a major component release (${current} -> X.0.0 with X > ${current_major}); got ${effective}"
+                errors=1
+                continue
+              fi
+            fi
+
+            # Without the marker the build would insert the new block above the
+            # file header instead of below it.
+            if ! grep -q '^<!-- changelog: release notes start -->$' "$component/CHANGELOG.md"; then
+              echo "::error::${component}/CHANGELOG.md is missing the '<!-- changelog: release notes start -->' marker; restore it after the intro line before compiling"
+              errors=1
+              continue
+            fi
+            # A hand-written UNRELEASED block means someone followed the old
+            # convention; its entries would be left out of the compiled block
+            # and out of the release notes extraction.
+            if grep -q '(Prowler UNRELEASED)' "$component/CHANGELOG.md"; then
+              echo "::error::${component}/CHANGELOG.md contains a hand-written '(Prowler UNRELEASED)' block; convert its entries to fragments in ${component}/changelog.d/ and delete the block before compiling"
+              errors=1
+              continue
+            fi
+
+            echo "${effective} ${mode}" > "${RUNNER_TEMP}/version-${component}.txt"
+            compiling="${compiling}${component} "
+          done
+          if [ "$errors" -ne 0 ]; then
+            exit 1
+          fi
+          if [ -z "$compiling" ]; then
+            echo "::error::Nothing to compile: no component has pending fragments to release"
+            exit 1
+          fi
+
+          body_file="${RUNNER_TEMP}/compile-changelogs-pr-body.md"
+          {
+            echo "### Description"
+            echo ""
+            echo "Compiles the pending changelog fragments into the per-component \`CHANGELOG.md\` files for Prowler v${PROWLER_VERSION}, replacing the manual stamping PR."
+            echo ""
+            echo "| Component | Version | Fragments consumed |"
+            echo "|---|---|---|"
+          } > "$body_file"
+
+          compiled_components=""
+          for component in prowler api ui mcp_server; do
+            if [ ! -f "${RUNNER_TEMP}/version-${component}.txt" ]; then
+              echo "Skipping ${component} (no pending fragments or held back)"
+              echo "| \`${component}\` | - | 0 |" >> "$body_file"
+              continue
+            fi
+            read -r version mode < "${RUNNER_TEMP}/version-${component}.txt"
+            version_label="$version"
+            if [ "$mode" = "auto" ]; then
+              version_label="${version} (auto)"
+            fi
+
+            count=$(pending_fragments "$component" | wc -l | tr -d ' ')
+            echo "Compiling ${component} ${version} (${count} fragments, ${mode} version)..."
+
+            # Captured before attribution renames them: these original paths are
+            # what the forward-sync deletes on master (backports copy fragments
+            # verbatim, so filenames match across branches).
+            pending_fragments "$component" > "${RUNNER_TEMP}/consumed-${component}.txt"
+            pre_lines=$(wc -l < "$component/CHANGELOG.md")
+
+            # Attribution must run before the build: towncrier renders the
+            # first dotted segment of each filename as the PR number.
+            python .github/scripts/changelog_attribution.py "$component"
+            towncrier build --config "$component/towncrier.toml" --version "$version" --name "Prowler v${PROWLER_VERSION}" --yes
+
+            # The build only inserts lines right after the marker, so the new
+            # stamped block is exactly the added lines following it. Captured
+            # for the forward-sync to master.
+            post_lines=$(wc -l < "$component/CHANGELOG.md")
+            delta=$((post_lines - pre_lines))
+            marker_line=$(grep -n -m1 '^<!-- changelog: release notes start -->$' "$component/CHANGELOG.md" | cut -d: -f1)
+            sed -n "$((marker_line + 1)),$((marker_line + delta))p" "$component/CHANGELOG.md" > "${RUNNER_TEMP}/block-${component}.md"
+
+            compiled_components="${compiled_components}${component} "
+            echo "| \`${component}\` | ${version_label} | ${count} |" >> "$body_file"
+          done
+          echo "COMPILED_COMPONENTS=${compiled_components}" >> "$GITHUB_ENV"
+
+          {
+            echo ""
+            echo "Review that no pending fragment was dropped (the diff must delete every consumed fragment) and that each new version block is correct, then squash-merge."
+            echo ""
+            echo "### License"
+            echo ""
+            echo "By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license."
+          } >> "$body_file"
+
+          echo "PR_BODY_FILE=${body_file}" >> "$GITHUB_ENV"
+
+      - name: Create compile PR
+        uses: peter-evans/create-pull-request@5f6978faf089d4d20b00c7766989d076bb2fc7f1 # v8.1.1
+        with:
+          token: ${{ secrets.PROWLER_BOT_ACCESS_TOKEN }}
+          commit-message: 'chore(changelog): v${{ env.PROWLER_VERSION }}'
+          branch: compile-changelogs-${{ env.PROWLER_VERSION }}
+          base: ${{ env.TARGET_BRANCH }}
+          title: 'chore(changelog): v${{ env.PROWLER_VERSION }}'
+          body-path: ${{ env.PR_BODY_FILE }}
+          author: prowler-bot <179230569+prowler-bot@users.noreply.github.com>
+          labels: |
+            no-changelog
+            skip-sync
+
+      # Patch compiles (target_branch = v5.X) leave master holding the consumed
+      # fragments and missing the new version block. This applies the equivalent
+      # change to master: insert the same stamped blocks under the marker and
+      # delete the consumed fragments, so the next minor compile cannot
+      # re-release entries that already shipped in the patch.
+      - name: Apply forward-sync to master
+        if: env.TARGET_BRANCH != 'master'
+        run: |
+          set -euo pipefail
+
+          git checkout -B master origin/master
+
+          version_key() {
+            local version="$1"
+            local major minor patch
+            IFS=. read -r major minor patch <<< "$version"
+            printf '%06d.%06d.%06d' "$((10#$major))" "$((10#$minor))" "$((10#$patch))"
+          }
+
+          release_from_heading() {
+            local heading="$1"
+            echo "$heading" | sed -E 's/^## \[[^]]+\] \(Prowler v?([0-9]+\.[0-9]+\.[0-9]+)\).*/\1/'
+          }
+
+          insert_changelog_block_ordered() {
+            local component="$1"
+            local block_file="$2"
+            local changelog="${component}/CHANGELOG.md"
+            local incoming_heading incoming_release incoming_key
+            local marker_line insertion_line duplicate_line total_lines
+            local line heading existing_release existing_key
+
+            marker_line=$(grep -n -m1 '^<!-- changelog: release notes start -->$' "$changelog" | cut -d: -f1)
+            incoming_heading=$(grep -m1 -E '^## \[[^]]+\] \(Prowler v?[0-9]+\.[0-9]+\.[0-9]+\)' "$block_file" || true)
+            if [ -z "$incoming_heading" ]; then
+              echo "::error::${block_file} does not contain a stamped Prowler release heading"
+              exit 1
+            fi
+
+            incoming_release=$(release_from_heading "$incoming_heading")
+            incoming_key=$(version_key "$incoming_release")
+            insertion_line=""
+            duplicate_line=""
+
+            while IFS=: read -r line heading; do
+              existing_release=$(release_from_heading "$heading")
+              existing_key=$(version_key "$existing_release")
+              if [[ "$incoming_key" == "$existing_key" ]]; then
+                duplicate_line="$line"
+                break
+              fi
+              if [[ "$incoming_key" > "$existing_key" ]]; then
+                insertion_line="$line"
+                break
+              fi
+            done < <(grep -n -E '^## \[[^]]+\] \(Prowler v?[0-9]+\.[0-9]+\.[0-9]+\)' "$changelog" || true)
+
+            if [ -n "$duplicate_line" ]; then
+              echo "::error::${changelog} already contains a block for Prowler v${incoming_release} at line ${duplicate_line}; refusing to insert a duplicate"
+              exit 1
+            fi
+            if [ -z "$insertion_line" ]; then
+              insertion_line=$(($(wc -l < "$changelog") + 1))
+            fi
+            if [ "$insertion_line" -le "$marker_line" ]; then
+              insertion_line=$((marker_line + 1))
+            fi
+
+            # The captured block window can be off by one blank line on either
+            # end (towncrier re-emits the blank after the marker), so strip the
+            # outer blank lines and pad exactly one on each side: the block
+            # must never glue to the marker above or the next heading below.
+            awk '
+              /[^[:space:]]/ { for (i = 0; i < pending; i++) print ""; pending = 0; print; started = 1; next }
+              started { pending++ }
+            ' "$block_file" > "${RUNNER_TEMP}/block-normalized.md"
+
+            total_lines=$(wc -l < "$changelog")
+            {
+              head -n "$((insertion_line - 1))" "$changelog"
+              if [ "$insertion_line" -gt 1 ] && [ -n "$(sed -n "$((insertion_line - 1))p" "$changelog")" ]; then
+                echo ""
+              fi
+              cat "${RUNNER_TEMP}/block-normalized.md"
+              if [ "$insertion_line" -le "$total_lines" ]; then
+                echo ""
+              fi
+              tail -n +"$insertion_line" "$changelog"
+            } > "${RUNNER_TEMP}/changelog.tmp"
+            mv "${RUNNER_TEMP}/changelog.tmp" "$changelog"
+
+            echo "::notice::Inserted ${component} changelog block for Prowler v${incoming_release} at line ${insertion_line}"
+          }
+
+          sync_body="${RUNNER_TEMP}/forward-sync-pr-body.md"
+          {
+            echo "### Description"
+            echo ""
+            echo "Forward-syncs the v${PROWLER_VERSION} compiled changelogs from \`${TARGET_BRANCH}\` to \`master\`: inserts the same stamped version blocks under the insertion marker and deletes the consumed fragments, so the next minor compile cannot re-release entries that already shipped in this patch. Opened automatically by the same run that opened the compile PR; review and squash-merge after it."
+            echo ""
+            echo "| Component | Fragments deleted on master | Skipped (only on ${TARGET_BRANCH}) |"
+            echo "|---|---|---|"
+          } > "$sync_body"
+
+          for component in $COMPILED_COMPONENTS; do
+            block_file="${RUNNER_TEMP}/block-${component}.md"
+            consumed_file="${RUNNER_TEMP}/consumed-${component}.txt"
+
+            if ! grep -qm1 '^<!-- changelog: release notes start -->$' "$component/CHANGELOG.md"; then
+              echo "::error::${component}/CHANGELOG.md on master is missing the insertion marker; cannot forward-sync"
+              exit 1
+            fi
+
+            deleted=0
+            skipped=0
+            while IFS= read -r fragment; do
+              if [ -z "$fragment" ]; then
+                continue
+              fi
+              if [ -f "$fragment" ]; then
+                git rm -q "$fragment"
+                deleted=$((deleted + 1))
+              else
+                echo "::notice::${fragment} does not exist on master (change landed only on ${TARGET_BRANCH}); skipping its deletion"
+                skipped=$((skipped + 1))
+              fi
+            done < "$consumed_file"
+
+            insert_changelog_block_ordered "$component" "$block_file"
+
+            echo "| \`${component}\` | ${deleted} | ${skipped} |" >> "$sync_body"
+          done
+
+          {
+            echo ""
+            echo "### License"
+            echo ""
+            echo "By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license."
+          } >> "$sync_body"
+
+          echo "SYNC_BODY_FILE=${sync_body}" >> "$GITHUB_ENV"
+
+      - name: Create forward-sync PR
+        if: env.TARGET_BRANCH != 'master'
+        uses: peter-evans/create-pull-request@5f6978faf089d4d20b00c7766989d076bb2fc7f1 # v8.1.1
+        with:
+          token: ${{ secrets.PROWLER_BOT_ACCESS_TOKEN }}
+          commit-message: 'chore(changelog): v${{ env.PROWLER_VERSION }} forward-sync to master'
+          branch: forward-sync-changelogs-${{ env.PROWLER_VERSION }}
+          base: master
+          title: 'chore(changelog): v${{ env.PROWLER_VERSION }} forward-sync to master'
+          body-path: ${{ env.SYNC_BODY_FILE }}
+          author: prowler-bot <179230569+prowler-bot@users.noreply.github.com>
+          labels: |
+            no-changelog
+            skip-sync

.github/workflows/conventional-commit.yml

@@ -26,7 +26,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/create-backport-label.yml

@@ -25,7 +25,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/docs-check-provider-cards.yml

@@ -0,0 +1,50 @@
+name: 'Docs: Check Provider Cards Snippet'
+
+on:
+  pull_request:
+    branches:
+      - 'master'
+      - 'v5.*'
+    paths:
+      - 'docs/user-guide/providers/**/getting-started-*.mdx'
+      - 'docs/scripts/generate_provider_cards.py'
+      - 'docs/snippets/provider-cards.mdx'
+      - 'api/src/backend/api/models.py'
+      - '.github/workflows/docs-check-provider-cards.yml'
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+permissions: {}
+
+jobs:
+  check-provider-cards:
+    if: github.repository == 'prowler-cloud/prowler'
+    runs-on: ubuntu-latest
+    timeout-minutes: 5
+    permissions:
+      contents: read
+
+    steps:
+      - name: Harden Runner
+        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        with:
+          egress-policy: block
+          allowed-endpoints: >
+            github.com:443
+
+      - name: Checkout repository
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        with:
+          persist-credentials: false
+
+      - name: Verify provider cards snippet is up to date
+        run: |
+          if ! python3 docs/scripts/generate_provider_cards.py; then
+            echo "::error::docs/snippets/provider-cards.mdx is out of sync with the provider getting-started pages or the API ProviderChoices enum."
+            echo "Run 'python3 docs/scripts/generate_provider_cards.py' locally and commit the regenerated snippet."
+            echo "--- diff ---"
+            git diff docs/snippets/provider-cards.mdx
+            exit 1
+          fi

.github/workflows/find-secrets.yml

@@ -25,14 +25,15 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           # We can't block as Trufflehog needs to verify secrets against vendors
           egress-policy: audit
-          # allowed-endpoints: >
-          #   github.com:443
-          #   ghcr.io:443
-          #   pkg-containers.githubusercontent.com:443
+          allowed-endpoints: >
+            github.com:443
+            ghcr.io:443
+            pkg-containers.githubusercontent.com:443
+            www.formbucket.com:443
 
       - name: Checkout repository
         uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

.github/workflows/helm-chart-checks.yml

@@ -33,7 +33,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/helm-chart-release.yml

@@ -26,7 +26,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/issue-lock-on-close.yml

@@ -22,7 +22,7 @@ jobs:
       issues: write
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/issue-triage.md

@@ -12,8 +12,8 @@ if: contains(toJson(github.event.issue.labels), 'status/needs-triage')
 
 timeout-minutes: 12
 
-rate-limit:
-  max: 5
+user-rate-limit:
+  max-runs-per-window: 5
   window: 60
 
 concurrency:
@@ -30,6 +30,12 @@ permissions:
 engine: copilot
 strict: false
 
+pre-steps:
+  - name: Harden the runner
+    uses: step-security/harden-runner@v2.20.0
+    with:
+      egress-policy: audit
+
 imports:
   - ../agents/issue-triage.md
 
@@ -108,7 +114,7 @@ Triage the following GitHub issue using the Prowler Issue Triage Agent persona.
 
 ## Sanitized Issue Content
 
-${{ needs.activation.outputs.text }}
+${{ steps.sanitized.outputs.text }}
 
 ## Instructions
 

.github/workflows/labeler.yml

@@ -27,7 +27,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -46,7 +46,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -93,7 +93,7 @@ jobs:
           fi
 
       - name: Add community label
-        if: steps.check_membership.outputs.is_member == 'false'
+        if: steps.check_membership.outputs.is_member == 'false' && github.event.pull_request.user.type != 'Bot'
         env:
           PR_NUMBER: ${{ github.event.pull_request.number }}
           GH_TOKEN: ${{ github.token }}

.github/workflows/markdown-lint.yml

@@ -26,7 +26,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@fa2e9d605c4eeb9fcad4c99c224cee0c6c7f3594 # v2.16.0
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/mcp-container-build-push.yml

@@ -45,7 +45,7 @@ jobs:
       contents: read
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
 
@@ -64,7 +64,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -106,7 +106,7 @@ jobs:
       packages: write
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -165,7 +165,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -206,7 +206,7 @@ jobs:
 
       - name: Install regctl
         if: always()
-        uses: regclient/actions/regctl-installer@da9319db8e44e8b062b3a147e1dfb2f574d41a03 # main
+        uses: regclient/actions/regctl-installer@9a2d4216180dbb3e2dccfa60d2dd4afd98e42ec5 # main
 
       - name: Cleanup intermediate architecture tags
         if: always()
@@ -227,7 +227,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -263,27 +263,3 @@ jobs:
           payload-file-path: "./.github/scripts/slack-messages/container-release-completed.json"
           step-outcome: ${{ steps.outcome.outputs.outcome }}
           update-ts: ${{ needs.notify-release-started.outputs.message-ts }}
-
-  trigger-deployment:
-    needs: [setup, container-build-push]
-    if: always() && github.event_name == 'push' && needs.setup.result == 'success' && needs.container-build-push.result == 'success'
-    runs-on: ubuntu-latest
-    timeout-minutes: 5
-    permissions:
-      contents: read
-
-    steps:
-      - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
-        with:
-          egress-policy: block
-          allowed-endpoints: >
-            api.github.com:443
-
-      - name: Trigger MCP deployment
-        uses: peter-evans/repository-dispatch@28959ce8df70de7be546dd1250a005dd32156697 # v4.0.1
-        with:
-          token: ${{ secrets.PROWLER_BOT_ACCESS_TOKEN }}
-          repository: ${{ secrets.CLOUD_DISPATCH }}
-          event-type: mcp-prowler-deployment
-          client-payload: '{"sha": "${{ github.sha }}", "short_sha": "${{ needs.setup.outputs.short-sha }}"}'

.github/workflows/mcp-container-checks.yml

@@ -33,7 +33,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -68,7 +68,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -102,6 +102,7 @@ jobs:
           files_ignore: |
             mcp_server/README.md
             mcp_server/CHANGELOG.md
+            mcp_server/changelog.d/**
 
       - name: Set up Docker Buildx
         if: steps.check-changes.outputs.any_changed == 'true'

.github/workflows/mcp-pypi-release.yml

@@ -29,7 +29,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -67,7 +67,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/mcp-security.yml

@@ -32,7 +32,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@a5ad31d6a139d249332a2605b85202e8c0b78450 # v2.19.1
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/nightly-arm64-container-builds.yml

@@ -48,7 +48,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -83,7 +83,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/pr-check-changelog.yml

@@ -19,6 +19,60 @@ concurrency:
 permissions: {}
 
 jobs:
+  test-changelog-attribution:
+    if: github.repository == 'prowler-cloud/prowler'
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    permissions:
+      contents: read
+
+    steps:
+      - name: Harden Runner
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
+        with:
+          egress-policy: block
+          allowed-endpoints: >
+            api.github.com:443
+            github.com:443
+            objects.githubusercontent.com:443
+            pypi.org:443
+            files.pythonhosted.org:443
+
+      - name: Checkout repository
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        with:
+          fetch-depth: 1
+          persist-credentials: false
+
+      - name: Fetch PR base ref for tj-actions/changed-files
+        env:
+          BASE_REF: ${{ github.event.pull_request.base.ref }}
+        run: git fetch --depth=1 origin "${BASE_REF}"
+
+      - name: Get changed files
+        id: changed-files
+        uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
+        with:
+          files: |
+            .github/scripts/changelog_attribution.py
+            .github/workflows/pr-check-changelog.yml
+            .github/workflows/compile-changelogs.yml
+            .github/towncrier/template.md.jinja
+            */towncrier.toml
+            tests/github/**
+
+      - name: Set up Python
+        if: steps.changed-files.outputs.any_changed == 'true'
+        uses: actions/setup-python@e797f83bcb11b83ae66e0230d6156d7c80228e7c # v6.0.0
+        with:
+          python-version: '3.12'
+
+      - name: Test changelog attribution
+        if: steps.changed-files.outputs.any_changed == 'true'
+        run: |
+          python3 -m pip install --user --disable-pip-version-check pytest==9.0.3 towncrier==25.8.0
+          python3 -m pytest tests/github
+
   check-changelog:
     if: contains(github.event.pull_request.labels.*.name, 'no-changelog') == false
     runs-on: ubuntu-latest
@@ -31,7 +85,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -62,53 +116,160 @@ jobs:
             uv.lock
             pyproject.toml
 
-      - name: Check for folder changes and changelog presence
+      - name: Check for folder changes and changelog fragment presence
         id: check-folders
         run: |
-          missing_changelogs=""
+          fragment_name_re='^[A-Za-z0-9][A-Za-z0-9._-]*\.(added|changed|deprecated|removed|fixed|security)(\.[0-9]+)?\.md$'
+          manual_pr_link_re='(\[\(#[0-9]+\)\]|\[#[0-9]+\]\(|\(#[0-9]+\)|github\.com/[^[:space:]/]+/[^[:space:]/]+/(pull|issues)/[0-9]+)'
+          folder_alt=$(echo "$MONITORED_FOLDERS" | tr ' ' '|')
+
+          missing_fragments=""
+          invalid_fragments=""
+          linked_fragments=""
+          handwritten_changelogs=""
+
+          all_changed=$(echo "${STEPS_CHANGED_FILES_OUTPUTS_ALL_CHANGED_FILES}" | tr ' ' '\n')
+          added=$(echo "${STEPS_CHANGED_FILES_OUTPUTS_ADDED_FILES}" | tr ' ' '\n')
+          added_or_renamed=$(printf '%s\n%s' "${STEPS_CHANGED_FILES_OUTPUTS_ADDED_FILES}" "${STEPS_CHANGED_FILES_OUTPUTS_RENAMED_FILES}" | tr ' ' '\n')
+          added_modified_or_renamed=$(printf '%s\n%s\n%s' "${STEPS_CHANGED_FILES_OUTPUTS_ADDED_FILES}" "${STEPS_CHANGED_FILES_OUTPUTS_MODIFIED_FILES}" "${STEPS_CHANGED_FILES_OUTPUTS_RENAMED_FILES}" | tr ' ' '\n')
+
+          # Returns success if the folder has a valid fragment added, modified, or renamed.
+          has_changelog_update() {
+            local folder="$1"
+            if echo "$added_modified_or_renamed" | grep "^${folder}/changelog.d/" | sed "s|^${folder}/changelog.d/||" | grep -qE "$fragment_name_re"; then
+              return 0
+            fi
+            return 1
+          }
 
           if [[ "${STEPS_CHANGED_FILES_OUTPUTS_ANY_CHANGED}" == "true" ]]; then
             # Check monitored folders
             for folder in $MONITORED_FOLDERS; do
-              # Get files changed in this folder
-              changed_in_folder=$(echo "${STEPS_CHANGED_FILES_OUTPUTS_ALL_CHANGED_FILES}" | tr ' ' '\n' | grep "^${folder}/" || true)
+              if echo "$all_changed" | grep -q "^${folder}/CHANGELOG.md$"; then
+                echo "Direct CHANGELOG.md edits are not allowed for ${folder}/"
+                handwritten_changelogs="${handwritten_changelogs}- \`${folder}/CHANGELOG.md\`"$'\n'
+              fi
+
+              changed_in_folder=$(echo "$all_changed" | grep "^${folder}/" | grep -v "^${folder}/CHANGELOG.md$" || true)
 
               if [ -n "$changed_in_folder" ]; then
                 echo "Detected changes in ${folder}/"
 
-                # Check if CHANGELOG.md was updated
-                if ! echo "$changed_in_folder" | grep -q "^${folder}/CHANGELOG.md$"; then
-                  echo "No changelog update found for ${folder}/"
-                  missing_changelogs="${missing_changelogs}- \`${folder}\`"$'\n'
+                if ! has_changelog_update "$folder"; then
+                  echo "No changelog fragment found for ${folder}/"
+                  missing_fragments="${missing_fragments}- \`${folder}\`"$'\n'
                 fi
               fi
             done
 
             # Check root-level dependency files (uv.lock, pyproject.toml)
             # These are associated with the prowler folder changelog
-            root_deps_changed=$(echo "${STEPS_CHANGED_FILES_OUTPUTS_ALL_CHANGED_FILES}" | tr ' ' '\n' | grep -E "^(uv\.lock|pyproject\.toml)$" || true)
+            root_deps_changed=$(echo "$all_changed" | grep -E "^(uv\.lock|pyproject\.toml)$" || true)
             if [ -n "$root_deps_changed" ]; then
               echo "Detected changes in root dependency files: $root_deps_changed"
-              # Check if prowler/CHANGELOG.md was already updated (might have been caught above)
-              prowler_changelog_updated=$(echo "${STEPS_CHANGED_FILES_OUTPUTS_ALL_CHANGED_FILES}" | tr ' ' '\n' | grep "^prowler/CHANGELOG.md$" || true)
-              if [ -z "$prowler_changelog_updated" ]; then
+              if ! has_changelog_update "prowler"; then
                 # Only add if prowler wasn't already flagged
-                if ! echo "$missing_changelogs" | grep -q "prowler"; then
-                  echo "No changelog update found for root dependency changes"
-                  missing_changelogs="${missing_changelogs}- \`prowler\` (root dependency files changed)"$'\n'
+                if ! echo "$missing_fragments" | grep -q "prowler"; then
+                  echo "No changelog fragment found for root dependency changes"
+                  missing_fragments="${missing_fragments}- \`prowler\` (root dependency files changed)"$'\n'
                 fi
               fi
             fi
+
+            # Validate the filename of every fragment added by this PR
+            added_fragments=$(echo "$added_or_renamed" | grep -E "^(${folder_alt})/changelog\.d/" || true)
+            for fragment in $added_fragments; do
+              name=$(basename "$fragment")
+              if [ "$name" = ".gitkeep" ] || [ "$name" = "README.md" ]; then
+                continue
+              fi
+              if ! echo "$name" | grep -qE "$fragment_name_re"; then
+                echo "Invalid fragment filename: $fragment"
+                invalid_fragments="${invalid_fragments}- \`${fragment}\`"$'\n'
+              fi
+            done
+
+            # Lint fragment content: the PR link is attached automatically at
+            # compile time, so a hand-written PR or issue link would be wrong
+            touched_fragments=$(echo "$added_modified_or_renamed" | grep -E "^(${folder_alt})/changelog\.d/" || true)
+            for fragment in $touched_fragments; do
+              name=$(basename "$fragment")
+              if [ "$name" = ".gitkeep" ] || [ "$name" = "README.md" ] || [ ! -f "$fragment" ]; then
+                continue
+              fi
+              if grep -qE "$manual_pr_link_re" "$fragment"; then
+                echo "Fragment contains a hand-written PR or issue link: $fragment"
+                linked_fragments="${linked_fragments}- \`${fragment}\`"$'\n'
+              fi
+            done
           fi
 
-          {
-            echo "missing_changelogs<<EOF"
-            echo -e "${missing_changelogs}"
-            echo "EOF"
-          } >> $GITHUB_OUTPUT
+          # Suggest a slug derived from the branch name for the bot comment
+          suggested_slug=$(echo "$HEAD_REF" | tr '[:upper:]' '[:lower:]' | sed 's|.*/||; s/[^a-z0-9._-]/-/g; s/^[^a-z0-9]*//')
+          if [ -z "$suggested_slug" ]; then
+            suggested_slug="my-change"
+          fi
+
+          fragment_help="A changelog fragment is a small Markdown file named \`<slug>.<type>.md\` under \`<component>/changelog.d/\`, where \`<type>\` is one of \`added\`, \`changed\`, \`deprecated\`, \`removed\`, \`fixed\` or \`security\`. Its content is the changelog entry text, without the PR link (added automatically at release time) and without a trailing period. For example:
+
+          \`\`\`
+          echo 'Entry text describing the change' > <component>/changelog.d/${suggested_slug}.fixed.md
+          \`\`\`
+
+          If this PR does not need a changelog entry, add the \`no-changelog\` label instead."
+
+          if [ -n "$missing_fragments" ] || [ -n "$invalid_fragments" ] || [ -n "$linked_fragments" ] || [ -n "$handwritten_changelogs" ]; then
+            comment_body=""
+            if [ -n "$missing_fragments" ]; then
+              comment_body="⚠️ **Changes detected in the following folders without a changelog fragment:**"$'\n\n'"${missing_fragments}"$'\n'
+            fi
+            if [ -n "$invalid_fragments" ]; then
+              comment_body="${comment_body}⚠️ **Changelog fragment filenames that do not follow the naming convention:**"$'\n\n'"${invalid_fragments}"$'\n'
+            fi
+            if [ -n "$linked_fragments" ]; then
+              comment_body="${comment_body}⚠️ **Changelog fragments containing a hand-written PR or issue link (remove it; the PR link is attached automatically at release time):**"$'\n\n'"${linked_fragments}"$'\n'
+            fi
+            if [ -n "$handwritten_changelogs" ]; then
+              comment_body="${comment_body}⚠️ **Direct \`CHANGELOG.md\` edits are not allowed in regular PRs:**"$'\n\n'"${handwritten_changelogs}"$'\n'
+            fi
+            comment_body="${comment_body}${fragment_help}"
+          else
+            comment_body="✅ All required changelog fragments are present."
+          fi
+
+          write_multiline_output() {
+            local name="$1"
+            local value="$2"
+            local delimiter
+
+            while true; do
+              delimiter="EOF_$(openssl rand -hex 16)"
+              if ! grep -qxF "$delimiter" <<< "$value"; then
+                break
+              fi
+            done
+
+            {
+              echo "${name}<<${delimiter}"
+              if [ -n "$value" ]; then
+                printf '%s\n' "$value"
+              fi
+              echo "${delimiter}"
+            } >> "$GITHUB_OUTPUT"
+          }
+
+          write_multiline_output "missing_fragments" "$missing_fragments"
+          write_multiline_output "invalid_fragments" "$invalid_fragments"
+          write_multiline_output "linked_fragments" "$linked_fragments"
+          write_multiline_output "handwritten_changelogs" "$handwritten_changelogs"
+          write_multiline_output "comment_body" "$comment_body"
         env:
           STEPS_CHANGED_FILES_OUTPUTS_ANY_CHANGED: ${{ steps.changed-files.outputs.any_changed }}
           STEPS_CHANGED_FILES_OUTPUTS_ALL_CHANGED_FILES: ${{ steps.changed-files.outputs.all_changed_files }}
+          STEPS_CHANGED_FILES_OUTPUTS_ADDED_FILES: ${{ steps.changed-files.outputs.added_files }}
+          STEPS_CHANGED_FILES_OUTPUTS_MODIFIED_FILES: ${{ steps.changed-files.outputs.modified_files }}
+          STEPS_CHANGED_FILES_OUTPUTS_RENAMED_FILES: ${{ steps.changed-files.outputs.renamed_files }}
+          HEAD_REF: ${{ github.event.pull_request.head.ref }}
 
       - name: Find existing changelog comment
         if: github.event.pull_request.head.repo.full_name == github.repository
@@ -128,14 +289,10 @@ jobs:
           edit-mode: replace
           body: |
             <!-- changelog-check -->
-            ${{ steps.check-folders.outputs.missing_changelogs != '' && format('⚠️ **Changes detected in the following folders without a corresponding update to the `CHANGELOG.md`:**
+            ${{ steps.check-folders.outputs.comment_body }}
 
-            {0}
-
-            Please add an entry to the corresponding `CHANGELOG.md` file to maintain a clear history of changes.', steps.check-folders.outputs.missing_changelogs) || '✅ All necessary `CHANGELOG.md` files have been updated.' }}
-
-      - name: Fail if changelog is missing
-        if: steps.check-folders.outputs.missing_changelogs != ''
+      - name: Fail if changelog fragment is missing or invalid
+        if: steps.check-folders.outputs.missing_fragments != '' || steps.check-folders.outputs.invalid_fragments != '' || steps.check-folders.outputs.linked_fragments != '' || steps.check-folders.outputs.handwritten_changelogs != ''
         run: |
-          echo "::error::Missing changelog updates in some folders"
+          echo "::error::Missing, invalid, or disallowed changelog updates"
           exit 1

.github/workflows/pr-check-compliance-mapping.yml

@@ -35,7 +35,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/pr-conflict-checker.yml

@@ -28,7 +28,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -37,8 +37,7 @@ jobs:
         with:
           ref: ${{ github.event.pull_request.head.sha }}
           fetch-depth: 1
-          # zizmor: ignore[artipacked]
-          persist-credentials: true # Required by tj-actions/changed-files to fetch PR branch
+          persist-credentials: false # No write token in the untrusted PR-head tree; public repo so base fetch/changed-files work unauthenticated
 
       - name: Fetch PR base ref for tj-actions/changed-files
         env:
@@ -50,6 +49,8 @@ jobs:
         uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
         with:
           files: '**'
+          safe_output: false # Raw paths (list read via env var, injection-safe); default escaping backslash-quotes chars like () and breaks the -f test
+          separator: "\n" # Newline-delimited so the reader tolerates spaces and glob chars in paths
 
       - name: Check for conflict markers
         id: conflict-check
@@ -59,19 +60,18 @@ jobs:
           CONFLICT_FILES=""
           HAS_CONFLICTS=false
 
-          # Check each changed file for conflict markers
-          for file in ${STEPS_CHANGED_FILES_OUTPUTS_ALL_CHANGED_FILES}; do
-            if [ -f "$file" ]; then
-              echo "Checking file: $file"
+          # Read newline-delimited paths so spaces/globs neither word-split nor glob-expand
+          while IFS= read -r file; do
+            [ -n "$file" ] || continue
+            [ -f "$file" ] || continue
+            echo "Checking file: $file"
 
-              # Look for conflict markers (more precise regex)
-              if grep -qE '^(<<<<<<<|=======|>>>>>>>)' "$file" 2>/dev/null; then
-                echo "Conflict markers found in: $file"
-                CONFLICT_FILES="${CONFLICT_FILES}- \`${file}\`"$'\n'
-                HAS_CONFLICTS=true
-              fi
+            if grep -qE '^(<<<<<<<|=======|>>>>>>>)' "$file" 2>/dev/null; then
+              echo "Conflict markers found in: $file"
+              CONFLICT_FILES="${CONFLICT_FILES}- \`${file}\`"$'\n'
+              HAS_CONFLICTS=true
             fi
-          done
+          done <<< "$STEPS_CHANGED_FILES_OUTPUTS_ALL_CHANGED_FILES"
 
           if [ "$HAS_CONFLICTS" = true ]; then
             echo "has_conflicts=true" >> $GITHUB_OUTPUT
@@ -88,18 +88,49 @@ jobs:
         env:
           STEPS_CHANGED_FILES_OUTPUTS_ALL_CHANGED_FILES: ${{ steps.changed-files.outputs.all_changed_files }}
 
+      - name: Check base-branch mergeability
+        id: merge-check
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          REPO: ${{ github.repository }}
+        run: |
+          MERGEABLE=null
+
+          # GitHub computes mergeability async, so .mergeable is null until ready; poll until resolved
+          for attempt in 1 2 3 4 5; do
+            MERGEABLE=$(gh api "repos/${REPO}/pulls/${PR_NUMBER}" --jq '.mergeable')
+            if [ "$MERGEABLE" != "null" ]; then
+              break
+            fi
+            echo "Attempt ${attempt}: mergeability not computed yet, retrying..."
+            sleep 3
+          done
+
+          # Keep 'unknown' distinct from 'clean' so we never assert a clean merge we could not confirm
+          case "$MERGEABLE" in
+            false) STATUS=conflict; echo "PR branch cannot be merged cleanly into its base branch" ;;
+            true)  STATUS=clean;    echo "PR branch merges cleanly into its base branch" ;;
+            *)     STATUS=unknown;  echo "::warning::Mergeability did not resolve after retries; leaving it undetermined" ;;
+          esac
+
+          echo "merge_status=${STATUS}" >> "$GITHUB_OUTPUT"
+
       - name: Manage conflict label
         env:
           GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
           PR_NUMBER: ${{ github.event.pull_request.number }}
           HAS_CONFLICTS: ${{ steps.conflict-check.outputs.has_conflicts }}
+          MERGE_STATUS: ${{ steps.merge-check.outputs.merge_status }}
         run: |
           LABEL_NAME="has-conflicts"
 
-          # Add or remove label based on conflict status
-          if [ "$HAS_CONFLICTS" = "true" ]; then
+          if [ "$HAS_CONFLICTS" = "true" ] || [ "$MERGE_STATUS" = "conflict" ]; then
             echo "Adding conflict label to PR #${PR_NUMBER}..."
             gh pr edit "$PR_NUMBER" --add-label "$LABEL_NAME" --repo ${{ github.repository }} || true
+          elif [ "$MERGE_STATUS" = "unknown" ]; then
+            # Don't drop the label on an undetermined merge state; a later run will settle it
+            echo "Mergeability undetermined; leaving label unchanged"
           else
             echo "Removing conflict label from PR #${PR_NUMBER}..."
             gh pr edit "$PR_NUMBER" --remove-label "$LABEL_NAME" --repo ${{ github.repository }} || true
@@ -121,20 +152,25 @@ jobs:
           edit-mode: replace
           body: |
             <!-- conflict-checker-comment -->
-            ${{ steps.conflict-check.outputs.has_conflicts == 'true' && '⚠️ **Conflict Markers Detected**' || '✅ **Conflict Markers Resolved**' }}
-
-            ${{ steps.conflict-check.outputs.has_conflicts == 'true' && format('This pull request contains unresolved conflict markers in the following files:
+            ${{ (steps.conflict-check.outputs.has_conflicts == 'true' || steps.merge-check.outputs.merge_status == 'conflict') && '⚠️ **Conflicts Detected**' || (steps.merge-check.outputs.merge_status == 'unknown' && 'ℹ️ **Conflict Check Incomplete**' || '✅ **No Conflicts**') }}
+            ${{ steps.conflict-check.outputs.has_conflicts == 'true' && format('
+            **Conflict markers** are present in the following files:
 
             {0}
-
-            Please resolve these conflicts by:
-            1. Locating the conflict markers: `<<<<<<<`, `=======`, and `>>>>>>>`
-            2. Manually editing the files to resolve the conflicts
-            3. Removing all conflict markers
-            4. Committing and pushing the changes', steps.conflict-check.outputs.conflict_files) || 'All conflict markers have been successfully resolved in this pull request.' }}
+            Resolve them by removing every `<<<<<<<`, `=======`, and `>>>>>>>` marker, then commit and push.', steps.conflict-check.outputs.conflict_files) || '' }}
+            ${{ steps.merge-check.outputs.merge_status == 'conflict' && '
+            **Merge conflict with the base branch.** This PR cannot be merged cleanly. Update your branch with the latest base (rebase or merge) and resolve the conflicts.' || '' }}
+            ${{ steps.merge-check.outputs.merge_status == 'unknown' && '
+            GitHub had not finished computing mergeability, so base-branch conflict status could not be verified on this run.' || '' }}
+            ${{ (steps.conflict-check.outputs.has_conflicts != 'true' && steps.merge-check.outputs.merge_status == 'clean') && '
+            No conflict markers, and the branch merges cleanly into its base.' || '' }}
 
       - name: Fail workflow if conflicts detected
-        if: steps.conflict-check.outputs.has_conflicts == 'true'
+        if: steps.conflict-check.outputs.has_conflicts == 'true' || steps.merge-check.outputs.merge_status == 'conflict'
+        env:
+          HAS_CONFLICTS: ${{ steps.conflict-check.outputs.has_conflicts }}
+          MERGE_STATUS: ${{ steps.merge-check.outputs.merge_status }}
         run: |
-          echo "::error::Workflow failed due to conflict markers detected in the PR"
+          [ "$HAS_CONFLICTS" = "true" ] && echo "::error::Conflict markers detected in changed files"
+          [ "$MERGE_STATUS" = "conflict" ] && echo "::error::PR branch has merge conflicts with the base branch"
           exit 1

.github/workflows/pr-merged.yml

@@ -26,7 +26,7 @@ jobs:
       contents: read
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -56,6 +56,6 @@ jobs:
               "PROWLER_PR_BODY": ${{ toJson(github.event.pull_request.body) }},
               "PROWLER_PR_URL": ${{ toJson(github.event.pull_request.html_url) }},
               "PROWLER_PR_MERGED_BY": "${{ github.event.pull_request.merged_by.login }}",
-              "PROWLER_PR_BASE_BRANCH": "${{ github.event.pull_request.base.ref }}",
-              "PROWLER_PR_HEAD_BRANCH": "${{ github.event.pull_request.head.ref }}"
+              "PROWLER_PR_BASE_BRANCH": ${{ toJson(github.event.pull_request.base.ref) }},
+              "PROWLER_PR_HEAD_BRANCH": ${{ toJson(github.event.pull_request.head.ref) }}
             }

.github/workflows/prepare-release.yml

@@ -29,7 +29,7 @@ jobs:
       pull-requests: write
     steps:
     - name: Harden the runner (Audit all outbound calls)
-      uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+      uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
       with:
         egress-policy: audit
 

.github/workflows/renovate-config-validate.yml

@@ -28,12 +28,13 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
             api.github.com:443
             github.com:443
+            raw.githubusercontent.com:443
             objects.githubusercontent.com:443
             codeload.github.com:443
             release-assets.githubusercontent.com:443
@@ -41,6 +42,7 @@ jobs:
             files.pythonhosted.org:443
             registry.npmjs.org:443
             nodejs.org:443
+            releases.astral.sh:443
 
       - name: Checkout repository
         uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

.github/workflows/sdk-check-duplicate-test-names.yml

@@ -25,7 +25,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/sdk-code-quality.yml

@@ -33,7 +33,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -55,6 +55,7 @@ jobs:
           files_ignore: |
             .github/**
             prowler/CHANGELOG.md
+            prowler/changelog.d/**
             docs/**
             permissions/**
             api/**

.github/workflows/sdk-codeql.yml

@@ -12,6 +12,7 @@ on:
       - '.github/workflows/sdk-codeql.yml'
       - '.github/codeql/sdk-codeql-config.yml'
       - '!prowler/CHANGELOG.md'
+      - '!prowler/changelog.d/**'
   pull_request:
     branches:
       - 'master'
@@ -23,6 +24,7 @@ on:
       - '.github/workflows/sdk-codeql.yml'
       - '.github/codeql/sdk-codeql-config.yml'
       - '!prowler/CHANGELOG.md'
+      - '!prowler/changelog.d/**'
   schedule:
     - cron: '00 12 * * *'
 
@@ -51,7 +53,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/sdk-container-build-push.yml

@@ -60,7 +60,7 @@ jobs:
       contents: read
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -98,7 +98,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -138,15 +138,18 @@ jobs:
     permissions:
       contents: read
       packages: write
+      id-token: write
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
             api.ecr-public.us-east-1.amazonaws.com:443
             public.ecr.aws:443
+            sts.amazonaws.com:443
+            sts.us-east-1.amazonaws.com:443
             registry-1.docker.io:443
             production.cloudflare.docker.com:443
             production.cloudfront.docker.com:443
@@ -173,14 +176,16 @@ jobs:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
 
-      - name: Login to Public ECR
-        uses: docker/login-action@4907a6ddec9925e35a0a9e82d7399ccc52663121 # v4.1.0
+      - name: Configure AWS credentials (OIDC)
+        uses: aws-actions/configure-aws-credentials@d979d5b3a71173a29b74b5b88418bfda9437d885 # v6.1.1
         with:
-          registry: public.ecr.aws
-          username: ${{ secrets.PUBLIC_ECR_AWS_ACCESS_KEY_ID }}
-          password: ${{ secrets.PUBLIC_ECR_AWS_SECRET_ACCESS_KEY }}
-        env:
-          AWS_REGION: ${{ env.AWS_REGION }}
+          aws-region: us-east-1
+          role-to-assume: ${{ secrets.PUBLIC_ECR_PUSH_ROLE_ARN }}
+
+      - name: Login to Public ECR
+        uses: aws-actions/amazon-ecr-login@d539f0932e70871a027e9d5a9d8fc38589180a64 # v2.1.6
+        with:
+          registry-type: public
 
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@4d04d5d9486b7bd6fa91e7baf45bbb4f8b9deedd # v4.0.0
@@ -206,10 +211,11 @@ jobs:
     runs-on: ubuntu-latest
     permissions:
       contents: read
+      id-token: write
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -221,6 +227,8 @@ jobs:
             github.com:443
             release-assets.githubusercontent.com:443
             api.ecr-public.us-east-1.amazonaws.com:443
+            sts.amazonaws.com:443
+            sts.us-east-1.amazonaws.com:443
 
 
       - name: Login to DockerHub
@@ -229,14 +237,16 @@ jobs:
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
 
-      - name: Login to Public ECR
-        uses: docker/login-action@4907a6ddec9925e35a0a9e82d7399ccc52663121 # v4.1.0
+      - name: Configure AWS credentials (OIDC)
+        uses: aws-actions/configure-aws-credentials@d979d5b3a71173a29b74b5b88418bfda9437d885 # v6.1.1
         with:
-          registry: public.ecr.aws
-          username: ${{ secrets.PUBLIC_ECR_AWS_ACCESS_KEY_ID }}
-          password: ${{ secrets.PUBLIC_ECR_AWS_SECRET_ACCESS_KEY }}
-        env:
-          AWS_REGION: ${{ env.AWS_REGION }}
+          aws-region: us-east-1
+          role-to-assume: ${{ secrets.PUBLIC_ECR_PUSH_ROLE_ARN }}
+
+      - name: Login to Public ECR
+        uses: aws-actions/amazon-ecr-login@d539f0932e70871a027e9d5a9d8fc38589180a64 # v2.1.6
+        with:
+          registry-type: public
 
       - name: Create and push manifests for push event
         if: github.event_name == 'push'
@@ -299,7 +309,7 @@ jobs:
 
       - name: Install regctl
         if: always()
-        uses: regclient/actions/regctl-installer@da9319db8e44e8b062b3a147e1dfb2f574d41a03 # main
+        uses: regclient/actions/regctl-installer@9a2d4216180dbb3e2dccfa60d2dd4afd98e42ec5 # main
 
       - name: Cleanup intermediate architecture tags
         if: always()
@@ -320,7 +330,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/sdk-container-checks.yml

@@ -35,7 +35,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -71,7 +71,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -94,6 +94,8 @@ jobs:
             _http._tcp.deb.debian.org:443
             powershellinfraartifacts-gkhedzdeaghdezhr.z01.azurefd.net:443
             get.trivy.dev:443
+            raw.githubusercontent.com:443
+            releases.astral.sh:443
 
       - name: Checkout repository
         uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
@@ -113,6 +115,7 @@ jobs:
             .github/workflows/sdk-container-checks.yml
           files_ignore: |
             prowler/CHANGELOG.md
+            prowler/changelog.d/**
             **/AGENTS.md
 
       - name: Set up Docker Buildx

.github/workflows/sdk-pypi-release.yml

@@ -28,7 +28,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -66,7 +66,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -102,7 +102,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/sdk-refresh-aws-services-regions.yml

@@ -27,7 +27,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/sdk-refresh-oci-regions.yml

@@ -25,7 +25,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/sdk-security.yml

@@ -37,7 +37,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -73,6 +73,7 @@ jobs:
             .github/scripts/osv-scan.sh
           files_ignore: |
             prowler/CHANGELOG.md
+            prowler/changelog.d/**
             **/AGENTS.md
 
       - name: Setup Python with uv

.github/workflows/sdk-tests.yml

@@ -33,7 +33,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -77,6 +77,7 @@ jobs:
           files_ignore: |
             .github/**
             prowler/CHANGELOG.md
+            prowler/changelog.d/**
             docs/**
             permissions/**
             api/**
@@ -614,6 +615,30 @@ jobs:
           flags: prowler-py${{ matrix.python-version }}-linode
           files: ./linode_coverage.xml
 
+      # E2E Networks Provider
+      - name: Check if E2E Networks files changed
+        if: steps.check-changes.outputs.any_changed == 'true'
+        id: changed-e2enetworks
+        uses: tj-actions/changed-files@9426d40962ed5378910ee2e21d5f8c6fcbf2dd96 # v47.0.6
+        with:
+          files: |
+            ./prowler/**/e2enetworks/**
+            ./tests/**/e2enetworks/**
+            ./uv.lock
+
+      - name: Run E2E Networks tests
+        if: steps.changed-e2enetworks.outputs.any_changed == 'true'
+        run: uv run pytest -n auto --cov=./prowler/providers/e2enetworks --cov-report=xml:e2enetworks_coverage.xml tests/providers/e2enetworks
+
+      - name: Upload E2E Networks coverage to Codecov
+        if: steps.changed-e2enetworks.outputs.any_changed == 'true'
+        uses: codecov/codecov-action@671740ac38dd9b0130fbe1cec585b89eea48d3de # v5.5.2
+        env:
+          CODECOV_TOKEN: ${{ secrets.CODECOV_TOKEN }}
+        with:
+          flags: prowler-py${{ matrix.python-version }}-e2enetworks
+          files: ./e2enetworks_coverage.xml
+
       # External Provider (dynamic loading)
       - name: Check if External Provider files changed
         if: steps.check-changes.outputs.any_changed == 'true'

.github/workflows/test-impact-analysis.yml

@@ -52,7 +52,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -73,7 +73,7 @@ jobs:
       - name: Setup Python
         uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
         with:
-          python-version: '3.12'
+          python-version: '3.12.13'
 
       - name: Install PyYAML
         run: pip install pyyaml

.github/workflows/ui-codeql.yml

@@ -10,6 +10,7 @@ on:
       - '.github/workflows/ui-codeql.yml'
       - '.github/codeql/ui-codeql-config.yml'
       - '!ui/CHANGELOG.md'
+      - '!ui/changelog.d/**'
   pull_request:
     branches:
       - 'master'
@@ -19,6 +20,7 @@ on:
       - '.github/workflows/ui-codeql.yml'
       - '.github/codeql/ui-codeql-config.yml'
       - '!ui/CHANGELOG.md'
+      - '!ui/changelog.d/**'
   schedule:
     - cron: '00 12 * * *'
 
@@ -47,7 +49,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/ui-container-build-push.yml

@@ -45,7 +45,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -64,7 +64,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -107,7 +107,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -160,7 +160,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -201,7 +201,7 @@ jobs:
 
       - name: Install regctl
         if: always()
-        uses: regclient/actions/regctl-installer@da9319db8e44e8b062b3a147e1dfb2f574d41a03 # main
+        uses: regclient/actions/regctl-installer@9a2d4216180dbb3e2dccfa60d2dd4afd98e42ec5 # main
 
       - name: Cleanup intermediate architecture tags
         if: always()
@@ -222,7 +222,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -258,27 +258,3 @@ jobs:
           payload-file-path: "./.github/scripts/slack-messages/container-release-completed.json"
           step-outcome: ${{ steps.outcome.outputs.outcome }}
           update-ts: ${{ needs.notify-release-started.outputs.message-ts }}
-
-  trigger-deployment:
-    needs: [setup, container-build-push]
-    if: always() && github.event_name == 'push' && needs.setup.result == 'success' && needs.container-build-push.result == 'success'
-    runs-on: ubuntu-latest
-    timeout-minutes: 5
-    permissions:
-      contents: read
-
-    steps:
-      - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
-        with:
-          egress-policy: block
-          allowed-endpoints: >
-            api.github.com:443
-
-      - name: Trigger UI deployment
-        uses: peter-evans/repository-dispatch@28959ce8df70de7be546dd1250a005dd32156697 # v4.0.1
-        with:
-          token: ${{ secrets.PROWLER_BOT_ACCESS_TOKEN }}
-          repository: ${{ secrets.CLOUD_DISPATCH }}
-          event-type: ui-prowler-deployment
-          client-payload: '{"sha": "${{ github.sha }}", "short_sha": "${{ needs.setup.outputs.short-sha }}"}'

.github/workflows/ui-container-checks.yml

@@ -33,7 +33,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -69,7 +69,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
@@ -102,6 +102,7 @@ jobs:
           files: ui/**
           files_ignore: |
             ui/CHANGELOG.md
+            ui/changelog.d/**
             ui/README.md
             ui/AGENTS.md
 

.github/workflows/ui-e2e-tests-v2.yml

@@ -14,6 +14,8 @@ on:
       - '.github/test-impact.yml'
       - 'ui/**'
       - 'api/**'  # API changes can affect UI E2E
+      - '!ui/CHANGELOG.md'
+      - '!api/CHANGELOG.md'
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
@@ -94,7 +96,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 
@@ -314,7 +316,7 @@ jobs:
       contents: read
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: audit
 

.github/workflows/ui-security.yml

@@ -32,7 +32,7 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >

.github/workflows/ui-tests.yml

@@ -31,14 +31,18 @@ jobs:
 
     steps:
       - name: Harden Runner
-        uses: step-security/harden-runner@ab7a9404c0f3da075243ca237b5fac12c98deaa5 # v2.19.3
+        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
         with:
           egress-policy: block
           allowed-endpoints: >
             github.com:443
             registry.npmjs.org:443
+            nodejs.org:443
             fonts.googleapis.com:443
             fonts.gstatic.com:443
+            api.iconify.design:443
+            api.simplesvg.com:443
+            api.unisvg.com:443
             api.github.com:443
             release-assets.githubusercontent.com:443
             cdn.playwright.dev:443
@@ -60,6 +64,7 @@ jobs:
             .github/workflows/ui-tests.yml
           files_ignore: |
             ui/CHANGELOG.md
+            ui/changelog.d/**
             ui/README.md
             ui/AGENTS.md
 

.gitignore

@@ -169,3 +169,7 @@ GEMINI.md
 
 # Claude Code
 .claude/*
+
+# Docker
+docker-compose.override.yml
+docker-compose-dev.override.yml

.pre-commit-config.yaml

@@ -72,13 +72,13 @@ repos:
         exclude: contrib
         priority: 30
 
-  ## PYTHON — SDK (prowler/, tests/, dashboard/, util/, scripts/)
+  ## PYTHON — SDK (prowler/, tests/, dashboard/, util/, scripts/, docs/scripts/)
   - repo: https://github.com/myint/autoflake
     rev: v2.3.3
     hooks:
       - id: autoflake
         name: "SDK - autoflake"
-        files: { glob: ["{prowler,tests,dashboard,util,scripts}/**/*.py"] }
+        files: { glob: ["{prowler,tests,dashboard,util,scripts,docs/scripts}/**/*.py"] }
         args: ["--in-place", "--remove-all-unused-imports", "--remove-unused-variable"]
         priority: 20
 
@@ -87,7 +87,7 @@ repos:
     hooks:
       - id: isort
         name: "SDK - isort"
-        files: { glob: ["{prowler,tests,dashboard,util,scripts}/**/*.py"] }
+        files: { glob: ["{prowler,tests,dashboard,util,scripts,docs/scripts}/**/*.py"] }
         args: ["--profile", "black"]
         stages: ["pre-commit"]
         priority: 20
@@ -97,7 +97,7 @@ repos:
     hooks:
       - id: black
         name: "SDK - black"
-        files: { glob: ["{prowler,tests,dashboard,util,scripts}/**/*.py"] }
+        files: { glob: ["{prowler,tests,dashboard,util,scripts,docs/scripts}/**/*.py"] }
         priority: 20
 
   - repo: https://github.com/pycqa/flake8
@@ -105,22 +105,49 @@ repos:
     hooks:
       - id: flake8
         name: "SDK - flake8"
-        files: { glob: ["{prowler,tests,dashboard,util,scripts}/**/*.py"] }
+        files: { glob: ["{prowler,tests,dashboard,util,scripts,docs/scripts}/**/*.py"] }
         args: ["--ignore=E266,W503,E203,E501,W605"]
         priority: 30
 
   ## PYTHON — API + MCP Server (ruff)
-  - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.15.11
+  # Run ruff through `uv run` against each project so prek uses the exact ruff
+  # version pinned in that project's uv.lock — the same version GitHub Actions
+  # runs via `uv run ruff`. This removes the drift between the local hooks and
+  # CI. api/ and mcp_server/ are separate uv projects, so they need separate
+  # hooks (each `uv run --project` resolves its own pinned ruff + config).
+  - repo: local
     hooks:
-      - id: ruff
-        name: "API + MCP - ruff check"
-        files: { glob: ["{api,mcp_server}/**/*.py"] }
-        args: ["--fix"]
+      - id: ruff-check-api
+        name: "API - ruff check"
+        entry: uv run --project ./api ruff check --fix
+        language: system
+        files: { glob: ["api/**/*.py"] }
         priority: 30
-      - id: ruff-format
-        name: "API + MCP - ruff format"
-        files: { glob: ["{api,mcp_server}/**/*.py"] }
+      - id: ruff-format-api
+        name: "API - ruff format"
+        entry: uv run --project ./api ruff format
+        language: system
+        files: { glob: ["api/**/*.py"] }
+        priority: 20
+      - id: ruff-check-mcp
+        name: "MCP - ruff check"
+        entry: uv run --project ./mcp_server ruff check --fix
+        language: system
+        files: { glob: ["mcp_server/**/*.py"] }
+        priority: 30
+      - id: ruff-format-mcp
+        name: "MCP - ruff format"
+        entry: uv run --project ./mcp_server ruff format
+        language: system
+        files: { glob: ["mcp_server/**/*.py"] }
+        priority: 20
+
+      - id: generate-provider-cards
+        name: "Docs - regenerate provider cards snippet"
+        entry: python3 docs/scripts/generate_provider_cards.py
+        language: system
+        files: { glob: ["docs/user-guide/providers/**/getting-started-*.mdx", "docs/scripts/generate_provider_cards.py", "docs/snippets/provider-cards.mdx", "api/src/backend/api/models.py"] }
+        pass_filenames: false
         priority: 20
 
   ## PYTHON — uv (API + SDK)
@@ -164,7 +191,7 @@ repos:
         entry: pylint --disable=W,C,R,E -j 0 -rn -sn
         language: system
         types: [python]
-        files: { glob: ["{prowler,tests,dashboard,util,scripts}/**/*.py"] }
+        files: { glob: ["{prowler,tests,dashboard,util,scripts,docs/scripts}/**/*.py"] }
         priority: 30
 
       - id: trufflehog

.trivyignore

@@ -15,21 +15,32 @@
 # neither vulnerable code path (Archive::Tar parsing or regex compilation of
 # attacker-controlled input) is reachable from Prowler. No Debian bookworm fix
 # is available yet.
-CVE-2026-42496 pkg:perl exp:2026-07-15
-CVE-2026-42496 pkg:perl-base exp:2026-07-15
-CVE-2026-42496 pkg:perl-modules-5.36 exp:2026-07-15
-CVE-2026-42496 pkg:libperl5.36 exp:2026-07-15
-CVE-2026-8376 pkg:perl exp:2026-07-15
-CVE-2026-8376 pkg:perl-base exp:2026-07-15
-CVE-2026-8376 pkg:perl-modules-5.36 exp:2026-07-15
-CVE-2026-8376 pkg:libperl5.36 exp:2026-07-15
+CVE-2026-42496 pkg:perl exp:2026-08-15
+CVE-2026-42496 pkg:perl-base exp:2026-08-15
+CVE-2026-42496 pkg:perl-modules-5.36 exp:2026-08-15
+CVE-2026-42496 pkg:libperl5.36 exp:2026-08-15
+CVE-2026-8376 pkg:perl exp:2026-08-15
+CVE-2026-8376 pkg:perl-base exp:2026-08-15
+CVE-2026-8376 pkg:perl-modules-5.36 exp:2026-08-15
+CVE-2026-8376 pkg:libperl5.36 exp:2026-08-15
+
+# CVE-2026-13221 - Perl regex trie overflow.
+# Packages: perl, perl-base, perl-modules-5.36, libperl5.36.
+# Why ignored: upstream confirms Perl 5.36.0 is not affected; the regression
+# was introduced after this version. Debian currently marks bookworm as
+# vulnerable, which causes Trivy to report a false positive.
+# Ref: https://github.com/Perl/perl5/issues/23388
+CVE-2026-13221 pkg:perl exp:2026-08-15
+CVE-2026-13221 pkg:perl-base exp:2026-08-15
+CVE-2026-13221 pkg:perl-modules-5.36 exp:2026-08-15
+CVE-2026-13221 pkg:libperl5.36 exp:2026-08-15
 
 # CVE-2025-7458 — SQLite integer overflow.
 # Package: libsqlite3-0.
 # Why ignored: transitive dependency of CPython's stdlib sqlite3 module. The
 # Prowler SDK does not open user-supplied SQLite databases; SQLite usage is
 # internal and bounded. No Debian bookworm fix is available.
-CVE-2025-7458 pkg:libsqlite3-0 exp:2026-07-15
+CVE-2025-7458 pkg:libsqlite3-0 exp:2026-08-15
 
 # CVE-2026-43185 — Linux kernel ksmbd signedness bug.
 # Package: linux-libc-dev.
@@ -37,7 +48,7 @@ CVE-2025-7458 pkg:libsqlite3-0 exp:2026-07-15
 # not a running kernel. Containers execute against the host kernel, so these
 # headers are inert at runtime. The upstream fix landed in kernel 7.0-rc2 and
 # has not been backported to Debian's 6.1 LTS line.
-CVE-2026-43185 pkg:linux-libc-dev exp:2026-07-15
+CVE-2026-43185 pkg:linux-libc-dev exp:2026-08-15
 
 # CVE-2023-45853 — zlib MiniZip integer overflow / heap overflow in
 # zipOpenNewFileInZip4_64.
@@ -49,8 +60,21 @@ CVE-2026-43185 pkg:linux-libc-dev exp:2026-07-15
 # zlib 1.3.1, available in Debian trixie (13); migrating the base image would
 # clear it fully.
 # Ref: https://security-tracker.debian.org/tracker/CVE-2023-45853
-CVE-2023-45853 pkg:zlib1g exp:2026-07-15
-CVE-2023-45853 pkg:zlib1g-dev exp:2026-07-15
+CVE-2023-45853 pkg:zlib1g exp:2026-08-15
+CVE-2023-45853 pkg:zlib1g-dev exp:2026-08-15
+
+# CVE-2026-55200 — libssh2 out-of-bounds write in ssh2_transport_read() due to
+# an unchecked packet_length field in transport.c (heap corruption, possible RCE).
+# Package: libssh2-1.
+# Why ignored: libssh2-1 is pulled in only as a transitive dependency of libcurl4
+# (installed in the SDK Dockerfile for the networking/PowerShell stack). The
+# vulnerable path is reached exclusively when libssh2 acts as an SSH/SCP/SFTP
+# client parsing transport packets from a server. Prowler never uses libcurl's
+# SSH/SCP/SFTP transports; it talks to cloud provider HTTPS endpoints only, so the
+# affected code is unreachable at runtime. Fixed upstream in libssh2 commit
+# 97acf3df (PR #2052); no Debian bookworm fix is available yet.
+# Ref: https://security-tracker.debian.org/tracker/CVE-2026-55200
+CVE-2026-55200 pkg:libssh2-1 exp:2026-08-15
 
 # --- API container image (api/Dockerfile) ---
 # The entries below are specific to the Prowler API image, which ships
@@ -65,13 +89,13 @@ CVE-2023-45853 pkg:zlib1g-dev exp:2026-07-15
 # at runtime. The vulnerable path requires parsing attacker-controlled XML with
 # the affected interpreter, which Prowler does not do with the system Python.
 # Full mitigation also needs libexpat >= 2.8.0; no Debian bookworm fix yet.
-CVE-2026-7210 pkg:python3.11 exp:2026-07-15
-CVE-2026-7210 pkg:python3.11-dev exp:2026-07-15
-CVE-2026-7210 pkg:python3.11-minimal exp:2026-07-15
-CVE-2026-7210 pkg:libpython3.11 exp:2026-07-15
-CVE-2026-7210 pkg:libpython3.11-dev exp:2026-07-15
-CVE-2026-7210 pkg:libpython3.11-minimal exp:2026-07-15
-CVE-2026-7210 pkg:libpython3.11-stdlib exp:2026-07-15
+CVE-2026-7210 pkg:python3.11 exp:2026-08-15
+CVE-2026-7210 pkg:python3.11-dev exp:2026-08-15
+CVE-2026-7210 pkg:python3.11-minimal exp:2026-08-15
+CVE-2026-7210 pkg:libpython3.11 exp:2026-08-15
+CVE-2026-7210 pkg:libpython3.11-dev exp:2026-08-15
+CVE-2026-7210 pkg:libpython3.11-minimal exp:2026-08-15
+CVE-2026-7210 pkg:libpython3.11-stdlib exp:2026-08-15
 
 # CVE-2026-33278 — Unbound DNSSEC validator use-after-free (DoS, possible RCE).
 # CVE-2026-42960 — Unbound DNS cache poisoning via promiscuous additional records.
@@ -81,5 +105,5 @@ CVE-2026-7210 pkg:libpython3.11-stdlib exp:2026-07-15
 # vulnerabilities require operating a live Unbound recursive DNSSEC validator
 # that processes attacker-influenced DNS responses. Prowler never starts an
 # Unbound resolver, so neither code path is reachable. No Debian bookworm fix yet.
-CVE-2026-33278 pkg:libunbound8 exp:2026-07-15
-CVE-2026-42960 pkg:libunbound8 exp:2026-07-15
+CVE-2026-33278 pkg:libunbound8 exp:2026-08-15
+CVE-2026-42960 pkg:libunbound8 exp:2026-08-15

AGENTS.md

@@ -114,6 +114,7 @@ When performing these actions, ALWAYS invoke the corresponding skill FIRST:
 | Review PR requirements: template, title conventions, changelog gate | `prowler-pr` |
 | Review changelog format and conventions | `prowler-changelog` |
 | Reviewing JSON:API compliance | `jsonapi` |
+| Reviewing Prowler UI components | `prowler-ui` |
 | Reviewing compliance framework PRs | `prowler-compliance-review` |
 | Running makemigrations or pgmakemigrations | `django-migration-psql` |
 | Syncing compliance framework with upstream catalog | `prowler-compliance` |

Dockerfile

@@ -1,12 +1,14 @@
-FROM python:3.12.13-slim-bookworm@sha256:76d4b7b6305788c6b4c6a19d6a22a3921bf802e9af4d5e1e5bd771208dba74bf AS build
+FROM python:3.12.13-slim-bookworm@sha256:8a7e7cc04fd3e2bd787f7f24e22d5d119aa590d429b50c95dfe12b3abe52f48b AS build
 
 LABEL maintainer="https://github.com/prowler-cloud/prowler"
 LABEL org.opencontainers.image.source="https://github.com/prowler-cloud/prowler"
 
 ARG POWERSHELL_VERSION=7.5.0
 ENV POWERSHELL_VERSION=${POWERSHELL_VERSION}
+# Opt out of PowerShell telemetry (Application Insights -> dc.services.visualstudio.com)
+ENV POWERSHELL_TELEMETRY_OPTOUT=1
 
-ARG TRIVY_VERSION=0.71.0
+ARG TRIVY_VERSION=0.71.2
 ENV TRIVY_VERSION=${TRIVY_VERSION}
 
 ARG ZIZMOR_VERSION=1.24.1
@@ -95,6 +97,18 @@ RUN uv sync --locked --compile-bytecode && \
 # Install PowerShell modules
 RUN .venv/bin/python prowler/providers/m365/lib/powershell/m365_powershell.py
 
+USER root
+
+# Remove build-only packages from the final image after Python dependencies are installed.
+RUN apt-get purge -y --auto-remove \
+    build-essential \
+    pkg-config \
+    libzstd-dev \
+    zlib1g-dev \
+    && rm -rf /var/lib/apt/lists/*
+
+USER prowler
+
 # Remove deprecated dash dependencies
 RUN pip uninstall dash-html-components -y && \
     pip uninstall dash-core-components -y

Makefile

@@ -45,18 +45,41 @@ coverage-html: ## Show Test Coverage
 	coverage html && \
 	open htmlcov/index.html
 
-##@ Linting
-format: ## Format Code
-	@echo "Running black..."
-	black .
+##@ Code Quality
+# `make` is the single entrypoint and mirrors CI exactly (uv run + same flags):
+#   SDK (prowler/, util/) -> flake8 + black + pylint
+#   API & MCP server      -> ruff (rules live in each project's pyproject.toml)
+# `format` applies fixes (incl. ruff's import/upgrade autofixes); `lint` only
+# verifies and is what CI gates on.
+.PHONY: format format-sdk format-api format-mcp lint lint-sdk lint-api lint-mcp
 
-lint: ## Lint Code
-	@echo "Running flake8..."
-	flake8 . --ignore=E266,W503,E203,E501,W605,E128 --exclude .venv,contrib
-	@echo "Running black... "
-	black --check .
-	@echo "Running pylint..."
-	pylint --disable=W,C,R,E -j 0 prowler util
+format: format-sdk format-api format-mcp ## Format & autofix all components (SDK, API, MCP)
+
+lint: lint-sdk lint-api lint-mcp ## Lint all components (SDK, API, MCP) — mirrors CI
+
+format-sdk: ## Format SDK code (black)
+	uv run black --exclude "\.venv|api|ui|skills|mcp_server" .
+
+lint-sdk: ## Lint SDK code (flake8, black --check, pylint)
+	uv run flake8 . --ignore=E266,W503,E203,E501,W605,E128 --exclude .venv,contrib,ui,api,skills,mcp_server
+	uv run black --exclude "\.venv|api|ui|skills|mcp_server" --check .
+	uv run pylint --disable=W,C,R,E -j 0 -rn -sn prowler/
+
+format-api: ## Format & autofix API code (ruff)
+	cd api && uv run ruff check . --exclude contrib --fix
+	cd api && uv run ruff format . --exclude contrib
+
+lint-api: ## Lint API code (ruff check + format --check)
+	cd api && uv run ruff check . --exclude contrib
+	cd api && uv run ruff format --check . --exclude contrib
+
+format-mcp: ## Format & autofix MCP server code (ruff)
+	cd mcp_server && uv run ruff check . --fix
+	cd mcp_server && uv run ruff format .
+
+lint-mcp: ## Lint MCP server code (ruff check + format --check)
+	cd mcp_server && uv run ruff check .
+	cd mcp_server && uv run ruff format --check .
 
 ##@ PyPI
 pypi-clean: ## Delete the distribution files

README.md

@@ -3,7 +3,7 @@
   <img align="center" alt="Prowler logo" src="https://github.com/prowler-cloud/prowler/blob/master/docs/img/prowler-logo-white.png#gh-dark-mode-only" width="50%" height="50%">
 </p>
 <p align="center">
-  <b><i>Prowler</b> is the Open Cloud Security Platform trusted by thousands to automate security and compliance in any cloud environment. With hundreds of ready-to-use checks and compliance frameworks, Prowler delivers real-time, customizable monitoring and seamless integrations, making cloud security simple, scalable, and cost-effective for organizations of any size.
+  <b><i>Prowler</b> is the Open Cloud Security Platform trusted by thousands to automate security and compliance in any cloud environment. With thousands of ready-to-use checks and compliance frameworks, Prowler delivers real-time, customizable monitoring and seamless integrations, making cloud security simple, scalable, and cost-effective for organizations of any size.
 </p>
 <p align="center">
 <b>Secure ANY cloud at AI Speed at <a href="https://prowler.com">prowler.com</i></b>
@@ -21,7 +21,7 @@
   <a href="https://pypi.python.org/pypi/prowler/"><img alt="Python Version" src="https://img.shields.io/pypi/pyversions/prowler.svg"></a>
   <a href="https://pypistats.org/packages/prowler"><img alt="PyPI Downloads" src="https://img.shields.io/pypi/dw/prowler.svg?label=downloads"></a>
   <a href="https://hub.docker.com/r/toniblyx/prowler"><img alt="Docker Pulls" src="https://img.shields.io/docker/pulls/toniblyx/prowler"></a>
-  <a href="https://gallery.ecr.aws/prowler-cloud/prowler"><img width="120" height=19" alt="AWS ECR Gallery" src="https://user-images.githubusercontent.com/3985464/151531396-b6535a68-c907-44eb-95a1-a09508178616.png"></a>
+  <a href="https://gallery.ecr.aws/prowler-cloud/prowler"><img width="120" height="19" alt="AWS ECR Gallery" src="https://user-images.githubusercontent.com/3985464/151531396-b6535a68-c907-44eb-95a1-a09508178616.png"></a>
   <a href="https://codecov.io/gh/prowler-cloud/prowler"><img alt="Codecov coverage" src="https://codecov.io/gh/prowler-cloud/prowler/graph/badge.svg?token=OflBGsdpDl"/></a>
   <a href="https://insights.linuxfoundation.org/project/prowler-cloud-prowler"><img alt="Linux Foundation insights health score" src="https://insights.linuxfoundation.org/api/badge/health-score?project=prowler-cloud-prowler"/></a>
 </p>
@@ -41,7 +41,7 @@
 
 # Description
 
-**Prowler** is the world’s most widely used _Open-Source Cloud Security Platform_ that automates security and compliance across **any cloud environment**. With hundreds of ready-to-use security checks, remediation guidance, and compliance frameworks, Prowler is built to _“Secure ANY Cloud at AI Speed”_. Prowler delivers **AI-driven**, **customizable**, and **easy-to-use** assessments, dashboards, reports, and integrations, making cloud security **simple**, **scalable**, and **cost-effective** for organizations of any size.
+**Prowler** is the world’s most widely used _Open-Source Cloud Security Platform_ that automates security and compliance across **any cloud environment**. With thousands of ready-to-use security checks, remediation guidance, and compliance frameworks, Prowler is built to _“Secure ANY Cloud at AI Speed”_. Prowler delivers **AI-driven**, **customizable**, and **easy-to-use** assessments, dashboards, reports, and integrations, making cloud security **simple**, **scalable**, and **cost-effective** for organizations of any size.
 
 Prowler includes hundreds of built-in controls to ensure compliance with standards and frameworks, including:
 
@@ -54,16 +54,16 @@ Prowler includes hundreds of built-in controls to ensure compliance with standar
 - **National Security Standards:** ENS (Spanish National Security Scheme) and KISA ISMS-P (Korean)
 - **Custom Security Frameworks:** Tailored to your needs
 
-## Prowler App / Prowler Cloud
+## Prowler Cloud & Prowler Local Server
 
-Prowler App / [Prowler Cloud](https://cloud.prowler.com/) is a web-based application that simplifies running Prowler across your cloud provider accounts. It provides a user-friendly interface to visualize the results and streamline your security assessments.
+[Prowler Cloud](https://cloud.prowler.com/) and Prowler Local Server, its self-hosted open-source version, are web applications that simplify running Prowler across your cloud provider accounts. They provide a user-friendly interface to visualize the results and streamline your security assessments.
 
-![Prowler App](docs/images/products/overview.png)
+![Prowler Cloud](docs/images/products/overview.png)
 ![Risk Pipeline](docs/images/products/risk-pipeline.png)
 ![Threat Map](docs/images/products/threat-map.png)
 
 
->For more details, refer to the [Prowler App Documentation](https://docs.prowler.com/projects/prowler-open-source/en/latest/#prowler-app-installation)
+>For more details, refer to the [Prowler Local Server documentation](https://docs.prowler.com/getting-started/installation/prowler-app)
 
 ## Prowler CLI
 
@@ -73,26 +73,45 @@ prowler <provider>
 ![Prowler CLI Execution](docs/img/short-display.png)
 
 
-## Prowler Dashboard
+## Prowler Local Dashboard
 
 ```console
 prowler dashboard
 ```
-![Prowler Dashboard](docs/images/products/dashboard.png)
+![Prowler Local Dashboard](docs/images/products/dashboard.png)
 
 
 ## Attack Paths
 
-Attack Paths automatically extends every completed AWS scan with a Neo4j graph that combines Cartography's cloud inventory with Prowler findings. The feature runs in the API worker after each scan and therefore requires:
+Attack Paths automatically extends every completed AWS scan with a graph that combines Cartography's cloud inventory with Prowler findings. The feature runs in the API worker after each scan.
 
-- An accessible Neo4j instance (the Docker Compose files already ships a `neo4j` service).
-- The following environment variables so Django and Celery can connect:
+Two graph backends are supported as the long-lived sink:
 
-  | Variable | Description | Default |
-  | --- | --- | --- |
-  | `NEO4J_HOST` | Hostname used by the API containers. | `neo4j` |
-  | `NEO4J_PORT` | Bolt port exposed by Neo4j. | `7687` |
-  | `NEO4J_USER` / `NEO4J_PASSWORD` | Credentials with rights to create per-tenant databases. | `neo4j` / `neo4j_password` |
+- **Neo4j** (default; the Docker Compose files already ship a `neo4j` service).
+- **Amazon Neptune** (cloud-managed; opt-in).
+
+Select the sink with `ATTACK_PATHS_SINK_DATABASE` (`neo4j` or `neptune`; default `neo4j`).
+
+> Note: Cartography ingestion always uses a temporary Neo4j database, regardless of the configured sink. The `NEO4J_*` variables below must remain set even when `ATTACK_PATHS_SINK_DATABASE=neptune`.
+
+### Neo4j sink
+
+| Variable | Description | Default |
+| --- | --- | --- |
+| `NEO4J_HOST` | Hostname used by the API containers. | `neo4j` |
+| `NEO4J_PORT` | Bolt port exposed by Neo4j. | `7687` |
+| `NEO4J_USER` / `NEO4J_PASSWORD` | Credentials with rights to create per-tenant databases. | `neo4j` / `neo4j_password` |
+
+### Neptune sink
+
+| Variable | Description | Default |
+| --- | --- | --- |
+| `NEPTUNE_WRITER_ENDPOINT` | Bolt host for the Neptune writer instance. Required when sink is `neptune`. | _empty_ |
+| `NEPTUNE_READER_ENDPOINT` | Optional reader endpoint for read-only queries. Falls back to the writer when unset. | _empty_ |
+| `NEPTUNE_PORT` | Bolt port exposed by Neptune. | `8182` |
+| `AWS_REGION` | Region the Neptune cluster lives in. Required when sink is `neptune`. | _empty_ |
+
+Neptune authenticates with SigV4 using the standard boto3 credential chain. The worker's IAM role (or `AWS_ACCESS_KEY_ID` / `AWS_SECRET_ACCESS_KEY`) supplies the credentials. There is no Neptune password variable.
 
 Every AWS provider scan will enqueue an Attack Paths ingestion job automatically. Other cloud providers will be added in future iterations.
 
@@ -102,29 +121,30 @@ Every AWS provider scan will enqueue an Attack Paths ingestion job automatically
 > For the most accurate and up-to-date information about checks, services, frameworks, and categories, visit [**Prowler Hub**](https://hub.prowler.com).
 
 
-| Provider | Checks | Services | [Compliance Frameworks](https://docs.prowler.com/projects/prowler-open-source/en/latest/tutorials/compliance/) | [Categories](https://docs.prowler.com/projects/prowler-open-source/en/latest/tutorials/misc/#categories) | Support | Interface |
+| Provider | Checks | Services | [Compliance Frameworks](https://docs.prowler.com/user-guide/compliance/tutorials/compliance) | [Categories](https://docs.prowler.com/user-guide/cli/tutorials/misc#categories) | Support | Interface |
 |---|---|---|---|---|---|---|
-| AWS | 600 | 84 | 44 | 18 | Official | UI, API, CLI |
-| Azure | 167 | 22 | 19 | 16 | Official | UI, API, CLI |
-| GCP | 102 | 18 | 17 | 12 | Official | UI, API, CLI |
-| Kubernetes | 83 | 7 | 7 | 11 | Official | UI, API, CLI |
-| GitHub | 24 | 3 | 1 | 5 | Official | UI, API, CLI |
-| M365 | 102 | 10 | 4 | 10 | Official | UI, API, CLI |
-| OCI | 51 | 14 | 4 | 10 | Official | UI, API, CLI |
-| Alibaba Cloud | 63 | 9 | 4 | 9 | Official | UI, API, CLI |
-| Cloudflare | 29 | 3 | 0 | 5 | Official | UI, API, CLI |
+| AWS | 615 | 86 | 47 | 19 | Official | UI, API, CLI |
+| Azure | 190 | 22 | 21 | 16 | Official | UI, API, CLI |
+| GCP | 109 | 20 | 19 | 12 | Official | UI, API, CLI |
+| Kubernetes | 90 | 7 | 8 | 11 | Official | UI, API, CLI |
+| GitHub | 24 | 3 | 2 | 5 | Official | UI, API, CLI |
+| M365 | 109 | 10 | 6 | 10 | Official | UI, API, CLI |
+| OCI | 52 | 14 | 5 | 10 | Official | UI, API, CLI |
+| Alibaba Cloud | 63 | 9 | 6 | 9 | Official | UI, API, CLI |
+| Cloudflare | 29 | 3 | 2 | 5 | Official | UI, API, CLI |
 | IaC | [See `trivy` docs.](https://trivy.dev/latest/docs/coverage/iac/) | N/A | N/A | N/A | Official | UI, API, CLI |
-| MongoDB Atlas | 10 | 3 | 0 | 8 | Official | UI, API, CLI |
+| MongoDB Atlas | 10 | 3 | 1 | 8 | Official | UI, API, CLI |
 | LLM | [See `promptfoo` docs.](https://www.promptfoo.dev/docs/red-team/plugins/) | N/A | N/A | N/A | Official | CLI |
 | Image | N/A | N/A | N/A | N/A | Official | CLI, API |
-| Google Workspace | 39 | 5 | 2 | 5 | Official | UI, API, CLI |
-| OpenStack | 34 | 5 | 0 | 9 | Official | UI, API, CLI |
-| Vercel | 26 | 6 | 0 | 8 | Official | UI, API, CLI |
-| Okta | 1 | 1 | 0 | 1 | Official | CLI |
-| Linode [Contact us](https://prowler.com/contact) | 10 | 3 | 0 | 4 | Unofficial | CLI |
-| Scaleway [Contact us](https://prowler.com/contact) | 1 | 1 | 0 | 1 | Unofficial | CLI |
-| StackIT [Contact us](https://prowler.com/contact) | 7 | 2 | 0 | 3 | Unofficial | CLI |
-| NHN | 6 | 2 | 1 | 0 | Unofficial | CLI |
+| Google Workspace | 65 | 11 | 3 | 6 | Official | UI, API, CLI |
+| OpenStack | 34 | 5 | 1 | 9 | Official | UI, API, CLI |
+| Vercel | 26 | 6 | 1 | 8 | Official | UI, API, CLI |
+| Okta | 29 | 8 | 2 | 2 | Official | UI, API, CLI |
+| Linode [Contact us](https://prowler.com/contact) | 10 | 3 | 1 | 4 | Unofficial | CLI |
+| E2E Networks [Contact us](https://prowler.com/contact) | 27 | 6 | 0 | 2 | Unofficial | CLI |
+| Scaleway [Contact us](https://prowler.com/contact) | 1 | 1 | 1 | 1 | Unofficial | CLI |
+| StackIT [Contact us](https://prowler.com/contact) | 7 | 2 | 1 | 3 | Unofficial | CLI |
+| NHN | 6 | 2 | 2 | 0 | Unofficial | CLI |
 
 > [!Note]
 > The numbers in the table are updated periodically.
@@ -140,11 +160,11 @@ Every AWS provider scan will enqueue an Attack Paths ingestion job automatically
 
 # 💻 Installation
 
-## Prowler App
+## Prowler Local Server
 
-Prowler App offers flexible installation methods tailored to various environments:
+Prowler Local Server offers flexible installation methods tailored to various environments:
 
-> For detailed instructions on using Prowler App, refer to the [Prowler App Usage Guide](https://docs.prowler.com/projects/prowler-open-source/en/latest/tutorials/prowler-app/).
+> For detailed instructions on using Prowler Local Server, refer to the [usage guide](https://docs.prowler.com/user-guide/tutorials/prowler-app).
 
 ### Docker Compose
 
@@ -177,7 +197,7 @@ docker compose up -d
 > [!WARNING]
 > 🔒 For a secure setup, the API auto-generates a unique key pair, `DJANGO_TOKEN_SIGNING_KEY` and `DJANGO_TOKEN_VERIFYING_KEY`, and stores it in `~/.config/prowler-api` (non-container) or the bound Docker volume in `_data/api` (container). Never commit or reuse static/default keys. To rotate keys, delete the stored key files and restart the API.
 
-Once configured, access the Prowler App at http://localhost:3000. Sign up using your email and password to get started.
+Once configured, access Prowler Local Server at http://localhost:3000. Sign up using your email and password to get started.
 
 ### Common Issues with Docker Pull Installation
 
@@ -249,7 +269,7 @@ pnpm run build
 pnpm start
 ```
 
-> Once configured, access the Prowler App at http://localhost:3000. Sign up using your email and password to get started.
+> Once configured, access Prowler Local Server at http://localhost:3000. Sign up using your email and password to get started.
 
 #### Pre-commit Hooks Setup
 
@@ -267,7 +287,7 @@ Prowler CLI is available as a project in [PyPI](https://pypi.org/project/prowler
 pip install prowler
 prowler -v
 ```
->For further guidance, refer to [https://docs.prowler.com](https://docs.prowler.com/projects/prowler-open-source/en/latest/#prowler-cli-installation)
+>For further guidance, refer to [https://docs.prowler.com](https://docs.prowler.com/getting-started/installation/prowler-cli)
 
 ### Containers
 
@@ -287,7 +307,7 @@ The container images are available here:
 - Prowler CLI:
     - [DockerHub](https://hub.docker.com/r/prowlercloud/prowler/tags)
     - [AWS Public ECR](https://gallery.ecr.aws/prowler-cloud/prowler)
-- Prowler App:
+- Prowler Local Server:
     - [DockerHub - Prowler UI](https://hub.docker.com/r/prowlercloud/prowler-ui/tags)
     - [DockerHub - Prowler API](https://hub.docker.com/r/prowlercloud/prowler-api/tags)
 
@@ -337,17 +357,55 @@ Full configuration, per-provider authentication, and SARIF examples: [Prowler Gi
 
 # ✏️ High level architecture
 
-## Prowler App
-**Prowler App** is composed of four key components:
+## Prowler Local Server
+**Prowler Local Server** is composed of four key components:
 
 - **Prowler UI**: A web-based interface, built with Next.js, providing a user-friendly experience for executing Prowler scans and visualizing results.
 - **Prowler API**: A backend service, developed with Django REST Framework, responsible for running Prowler scans and storing the generated results.
 - **Prowler SDK**: A Python SDK designed to extend the functionality of the Prowler CLI for advanced capabilities.
 - **Prowler MCP Server**: A Model Context Protocol server that provides AI tools for Lighthouse, the AI-powered security assistant. This is a critical dependency for Lighthouse functionality.
 
-![Prowler App Architecture](docs/images/products/prowler-app-architecture.png)
+```mermaid
+flowchart TB
+    user([User / Security Team])
+    cli([Prowler CLI])
 
-<!-- Diagram source: docs/images/products/prowler-app-architecture.mmd — edit there, re-render at https://mermaid.live, and replace the PNG. -->
+    subgraph APP["Prowler Local Server"]
+        ui["Prowler UI<br/>(Next.js)"]
+        api["Prowler API<br/>(Django REST Framework)"]
+        worker["API Worker<br/>(Celery)"]
+        beat["API Scheduler<br/>(Celery Beat)"]
+        mcp["Prowler MCP Server<br/>(Lighthouse AI tools)"]
+    end
+
+    sdk["Prowler SDK<br/>(Python)"]
+
+    subgraph DATA["Data Layer"]
+        pg[("PostgreSQL")]
+        valkey[("Valkey / Redis")]
+        neo4j[("Neo4j")]
+    end
+
+    providers["Providers"]
+
+    user --> ui
+    user --> cli
+    ui -->|REST| api
+    ui -->|MCP HTTP| mcp
+    mcp -->|REST| api
+    api --> pg
+    api --> valkey
+    beat -->|enqueue jobs| valkey
+    valkey -->|dispatch| worker
+    worker --> pg
+    worker -->|Attack Paths| neo4j
+    worker -->|invokes| sdk
+    cli --> sdk
+
+    sdk --> providers
+```
+
+<!-- Diagram source: docs/images/products/prowler-app-architecture.mmd — keep this inline block, the docs page getting-started/products/prowler-app.mdx, and the .mmd file in sync. -->
 
 
 ## Prowler CLI

api/CHANGELOG.md

@@ -2,7 +2,102 @@
 
 All notable changes to the **Prowler API** are documented in this file.
 
-## [1.32.0] (Prowler UNRELEASED)
+<!-- changelog: release notes start -->
+
+## [1.35.0] (Prowler v5.34.0)
+
+### 🐞 Fixed
+
+- `rls_transaction` now falls back directly to the primary DB for connection-level mid-query read replica failures via `execute_wrapper`, reducing non-streaming read crashes during replica recovery [(#10379)](https://github.com/prowler-cloud/prowler/pull/10379)
+- RBAC permission gates now combine permissions from every role assigned to a user in the active tenant [(#11979)](https://github.com/prowler-cloud/prowler/pull/11979)
+- `attack-paths-cleanup-stale-scans` now retries worker pings and checks recent scan activity before failing scans and removing temporary databases [(#11986)](https://github.com/prowler-cloud/prowler/pull/11986)
+
+### 🔐 Security
+
+- User role relationship updates are limited to the active tenant to preserve role assignments in other tenants [(#11903)](https://github.com/prowler-cloud/prowler/pull/11903)
+- `api` container image removes the unused Debian `libxml2` runtime package and scopes the `CVE-2026-13221` Trivy exception to unaffected Perl 5.36 packages [(#11991)](https://github.com/prowler-cloud/prowler/pull/11991)
+
+---
+
+## [1.34.2] (Prowler v5.33.2)
+
+### 🐞 Fixed
+
+- Attack Paths graph mutations now retry transient Neptune concurrency and deadline failures, while Neo4j mutations use managed transaction retries [(#11968)](https://github.com/prowler-cloud/prowler/pull/11968)
+- Attack Paths scans now use bounded child node identifiers for normalized list values in Neo4j and Neptune, preventing Neo4j RANGE index key size failures [(#11969)](https://github.com/prowler-cloud/prowler/pull/11969)
+- `scan-summary` aggregation now upserts summaries in deterministic conflict-key order, preventing PostgreSQL deadlocks during concurrent reaggregation [(#11971)](https://github.com/prowler-cloud/prowler/pull/11971)
+
+---
+
+## [1.34.1] (Prowler v5.33.1)
+
+### 🐞 Fixed
+
+- Session tokens are rejected after account password updates [(#11914)](https://github.com/prowler-cloud/prowler/pull/11914)
+- Jira dispatch task results now surface user-facing Jira failure messages [(#11925)](https://github.com/prowler-cloud/prowler/pull/11925)
+- AWS Attack Paths privilege escalation queries no longer fail on Neo4j with `Aggregation column contains implicit grouping expressions` [(#11939)](https://github.com/prowler-cloud/prowler/pull/11939)
+
+### 🔐 Security
+
+- OpenAI-compatible Lighthouse provider base URLs are restricted before connection checks [(#11940)](https://github.com/prowler-cloud/prowler/pull/11940)
+- `LIGHTHOUSE_AI_OPENAI_COMPATIBLE_ALLOWED_HOSTS` environment variable to allow internal hosts as OpenAI-compatible Lighthouse AI base URLs [(#11942)](https://github.com/prowler-cloud/prowler/pull/11942)
+
+---
+
+## [1.34.0] (Prowler v5.33.0)
+
+### 🚀 Added
+
+- Compliance PDF reports no longer require provider credentials: findings are enriched from the provider metadata stored in the database, so reports generate even after the provider secret is deleted or its credentials become invalid [(#11845)](https://github.com/prowler-cloud/prowler/pull/11845)
+
+### 🐞 Fixed
+
+- Provider scans now queue behind active provider scans instead of dispatching concurrently, and resource failed-finding counters retry database conflicts with stable row locking [(#11848)](https://github.com/prowler-cloud/prowler/pull/11848)
+
+---
+
+## [1.33.1] (Prowler v5.32.1)
+
+### 🐞 Fixed
+
+- Attack Paths: Scan rows now have database defaults for `is_migrated` and `sink_backend` so `scan-perform-scheduled` inserts survive deploy skew [(#11826)](https://github.com/prowler-cloud/prowler/pull/11826)
+- Invited users now keep their invitation context when completing authentication with Google, GitHub, or SAML, so the invitation is accepted during login [(#11752)](https://github.com/prowler-cloud/prowler/pull/11752)
+
+### 🔐 Security
+
+- User profile updates now allow users to update their own account while requiring user-management permissions to update other users in the same tenant [(#11792)](https://github.com/prowler-cloud/prowler/pull/11792)
+- Kubernetes provider credentials now reject kubeconfigs using `exec` authentication in Prowler Cloud, preventing user-supplied commands from running on Cloud workers [(#11753)](https://github.com/prowler-cloud/prowler/pull/11753)
+
+---
+
+## [1.33.0] (Prowler v5.32.0)
+
+### 🚀 Added
+
+- Timestamp precision support for `/api/v1/findings` `inserted_at` and `updated_at` filters [(#11754)](https://github.com/prowler-cloud/prowler/pull/11754)
+
+### 🔄 Changed
+
+- Attack Paths: AWS Neptune is now supported as a persistent sink database, selectable via `ATTACK_PATHS_SINK_DATABASE=neptune` (default `neo4j`), Cartography's (bumped to 0.138.1) per-scan ingest database stays on Neo4j [(#11524)](https://github.com/prowler-cloud/prowler/pull/11524)
+- Attack Paths: Scan task now checks the ingest Neo4j database and configured graph sink before starting graph ingestion [(#11743)](https://github.com/prowler-cloud/prowler/pull/11743)
+- Disable PowerShell telemetry in the API container image [(#11746)](https://github.com/prowler-cloud/prowler/pull/11746)
+
+### 🐞 Fixed
+
+- Attack Paths: Provider graph cleanup now deletes Neo4j and Neptune relationships in directed batches before deleting nodes [(#11755)](https://github.com/prowler-cloud/prowler/pull/11755)
+- `scan-perform` no longer reports an error when a provider is deleted during a running scan [(#11696)](https://github.com/prowler-cloud/prowler/pull/11696)
+
+---
+
+## [1.32.1] (Prowler v5.31.1)
+
+### 🐞 Fixed
+
+- API key auth no longer mutates `TenantAPIKey.objects` during admin DB lookups [(#11686)](https://github.com/prowler-cloud/prowler/pull/11686)
+
+---
+
+## [1.32.0] (Prowler v5.31.0)
 
 ### 🚀 Added
 
@@ -17,6 +112,7 @@ All notable changes to the **Prowler API** are documented in this file.
 - Sentry now drops ASGI's `RequestAborted` errors from health-check probe disconnects on `/health/live` [(#11632)](https://github.com/prowler-cloud/prowler/pull/11632)
 - Gunicorn keep-alive timeout now exceeds the load balancer idle timeout, stopping 502s from reused connections [(#11647)](https://github.com/prowler-cloud/prowler/pull/11647)
 - API runs under the Uvicorn worker so keep-alive outlives the load balancer idle timeout, fixing Gunicorn's intermittent 502s [(#11663)](https://github.com/prowler-cloud/prowler/pull/11663)
+- SAML logins no longer wipe a user's roles when the IdP does not send the `userType` attribute; existing roles are kept, and when `userType` names a role that does not exist it is now created with read-only access (visibility over all providers, no management permissions) instead of no permissions at all [(#11520)](https://github.com/prowler-cloud/prowler/pull/11520)
 
 ### 🐞 Fixed
 
@@ -68,7 +164,6 @@ All notable changes to the **Prowler API** are documented in this file.
 ### 🔄 Changed
 
 - Allowlisted idempotent background tasks are no longer lost when a worker is stopped or crashes mid-task; tasks with external side effects are marked terminal instead of blindly re-running [(#11416)](https://github.com/prowler-cloud/prowler/pull/11416)
-- SAML logins no longer wipe a user's roles when the IdP does not send the `userType` attribute; existing roles are kept, and when `userType` names a role that does not exist it is now created with read-only access (visibility over all providers, no management permissions) instead of no permissions at all [(#11520)](https://github.com/prowler-cloud/prowler/pull/11520)
 
 ### 🐞 Fixed
 

api/Dockerfile

@@ -1,11 +1,13 @@
-FROM python:3.12.13-slim-bookworm@sha256:76d4b7b6305788c6b4c6a19d6a22a3921bf802e9af4d5e1e5bd771208dba74bf AS build
+FROM python:3.12.13-slim-bookworm@sha256:8a7e7cc04fd3e2bd787f7f24e22d5d119aa590d429b50c95dfe12b3abe52f48b AS build
 
 LABEL maintainer="https://github.com/prowler-cloud/api"
 
 ARG POWERSHELL_VERSION=7.5.0
 ENV POWERSHELL_VERSION=${POWERSHELL_VERSION}
+# Opt out of PowerShell telemetry (Application Insights -> dc.services.visualstudio.com)
+ENV POWERSHELL_TELEMETRY_OPTOUT=1
 
-ARG TRIVY_VERSION=0.71.0
+ARG TRIVY_VERSION=0.71.2
 ENV TRIVY_VERSION=${TRIVY_VERSION}
 
 ARG ZIZMOR_VERSION=1.24.1
@@ -102,6 +104,24 @@ RUN uv sync --locked --no-install-project && \
 
 RUN .venv/bin/python .venv/lib/python3.12/site-packages/prowler/providers/m365/lib/powershell/m365_powershell.py
 
+USER root
+
+# Remove build-only packages from the final image after Python dependencies are installed.
+RUN apt-get purge -y --auto-remove \
+    gcc \
+    g++ \
+    make \
+    libxml2-dev \
+    libxmlsec1-dev \
+    libxmlsec1-openssl \
+    pkg-config \
+    libtool \
+    libxslt1-dev \
+    python3-dev \
+    && rm -rf /var/lib/apt/lists/*
+
+USER prowler
+
 COPY --chown=prowler:prowler src/backend/ ./backend/
 COPY --chown=prowler:prowler docker-entrypoint.sh ./docker-entrypoint.sh
 

api/changelog.d/README.md

@@ -0,0 +1,10 @@
+# Changelog fragments
+
+Each PR adds one small file here instead of editing `CHANGELOG.md` directly, so concurrent PRs never conflict.
+
+- Filename: `<slug>.<type>.md`, e.g. `my-new-check.added.md` (slug is free-form: letters, digits, `.`, `_`, `-`)
+- `<type>` is one of: `added`, `changed`, `deprecated`, `removed`, `fixed`, `security`
+- Content: one line with the changelog entry text, without the PR link and without a trailing period (the PR link is attached automatically at release time)
+- A PR adds as many fragment files as entries it needs, freely mixing types (one file per entry); same-type entries just use different slugs
+
+Fragments are compiled into `CHANGELOG.md` when a release is prepared. Full conventions: `skills/prowler-changelog/SKILL.md`.

api/changelog.d/attack-paths-scan-reliability.fixed.md

@@ -0,0 +1 @@
+Attack Paths scans handle provider deletion races cleanly, detect stale tasks after 16 hours, use backend-specific graph synchronization batches, and report exhausted Neptune write retries with the original database error

api/changelog.d/attack-paths-scan-time-limit.fixed.md

@@ -0,0 +1 @@
+`attack-paths-scan-perform` Celery tasks now use the configurable long-task time limits instead of the six-hour defaults

api/changelog.d/jira-site-name-validation.security.md

@@ -0,0 +1 @@
+Jira integration credentials only accept bare Atlassian site names containing letters, numbers, and hyphens

api/changelog.d/social-account-linking.security.md

@@ -0,0 +1 @@
+Social account linking requires a verified matching email from both the identity provider and the existing user account without sending account connection notifications

api/pyproject.toml

@@ -14,7 +14,7 @@ dev = [
   "pytest-env==1.1.3",
   "pytest-randomly==3.15.0",
   "pytest-xdist==3.6.1",
-  "ruff==0.5.0",
+  "ruff==0.15.11",
   "tqdm==4.67.1",
   "vulture==2.14",
   "prek==0.3.9"
@@ -58,7 +58,7 @@ dependencies = [
   "matplotlib (==3.10.8)",
   "reportlab (==4.4.10)",
   "neo4j (==6.1.0)",
-  "cartography (==0.135.0)",
+  "cartography (==0.138.1)",
   "gevent (==25.9.1)",
   "werkzeug (==3.1.7)",
   "sqlparse (==0.5.5)",
@@ -71,7 +71,24 @@ name = "prowler-api"
 package-mode = false
 # Needed for the SDK compatibility
 requires-python = ">=3.11,<3.13"
-version = "1.32.0"
+version = "1.36.0"
+
+# Shared ruff baseline (kept in sync with mcp_server/pyproject.toml).
+# target-version tracks this project's lowest supported Python.
+[tool.ruff]
+src = ["src"]
+target-version = "py311"
+
+[tool.ruff.lint]
+# Defaults (E4/E7/E9, F) plus import sorting, modern-syntax upgrades, and
+# comprehension lints — all mechanically auto-fixable. flake8-bugbear (B) is a
+# good next step but needs manual cleanup (e.g. B904 raise-from), so it is left
+# out of the shared baseline for now.
+extend-select = [
+  "I",  # isort — import ordering (prek's isort hook covers only the SDK)
+  "UP",  # pyupgrade — modern syntax for the min supported Python
+  "C4"  # flake8-comprehensions
+]
 
 [tool.uv]
 # Transitive pins matching master to avoid silent drift; bump deliberately.
@@ -176,7 +193,7 @@ constraint-dependencies = [
   "blinker==1.9.0",
   "boto3==1.40.61",
   "botocore==1.40.61",
-  "cartography==0.135.0",
+  "cartography==0.138.1",
   "celery==5.6.2",
   "certifi==2026.1.4",
   "cffi==2.0.0",
@@ -201,7 +218,6 @@ constraint-dependencies = [
   "debugpy==1.8.20",
   "decorator==5.2.1",
   "defusedxml==0.7.1",
-  "detect-secrets==1.5.0",
   "dill==0.4.1",
   "distro==1.9.0",
   "dj-rest-auth==7.0.1",
@@ -284,6 +300,7 @@ constraint-dependencies = [
   "jsonschema==4.23.0",
   "jsonschema-specifications==2025.9.1",
   "keystoneauth1==5.13.0",
+  "kingfisher-bin==1.104.0",
   "kiwisolver==1.4.9",
   "knack==0.11.0",
   "kombu==5.6.2",
@@ -393,7 +410,7 @@ constraint-dependencies = [
   "rpds-py==0.30.0",
   "rsa==4.9.1",
   "ruamel-yaml==0.19.1",
-  "ruff==0.5.0",
+  "ruff==0.15.11",
   "s3transfer==0.14.0",
   "scaleway==2.10.3",
   "scaleway-core==2.10.3",
@@ -430,7 +447,7 @@ constraint-dependencies = [
   "wcwidth==0.5.3",
   "websocket-client==1.9.0",
   "werkzeug==3.1.7",
-  "workos==6.0.4",
+  "workos==6.0.8",
   "wrapt==1.17.3",
   "xlsxwriter==3.2.9",
   "xmlsec==1.3.17",
@@ -441,8 +458,13 @@ constraint-dependencies = [
   "zope-interface==8.2",
   "zstd==1.5.7.3"
 ]
-# prowler@master needs okta==3.4.2; cartography 0.135.0 declares okta<1.0.0 for an
-# integration prowler does not import.
+# prowler@master needs okta==3.4.2, but cartography 0.138.1 requires okta<1.0.0.
+# Attack Paths does not ingest Okta today, so override the Cartography
+# dependency to the Prowler pin.
+#
+# prowler@master needs azure-mgmt-containerservice==34.1.0, but cartography
+# 0.138.1 requires azure-mgmt-containerservice>=41.0.0. Attack Paths does not
+# ingest Azure today, so override the Cartography dependency to the Prowler pin.
 #
 # prowler@master hard-pins microsoft-kiota-abstractions==1.9.2 in [project.dependencies].
 # The microsoft-kiota-http security bump to 1.9.9 (GHSA-7j59-v9qr-6fq9) requires
@@ -458,6 +480,7 @@ constraint-dependencies = [
 # that request pyjwt[crypto] and leave cryptography (needed for RS256) only transitive.
 override-dependencies = [
   "okta==3.4.2",
+  "azure-mgmt-containerservice==34.1.0",
   "microsoft-kiota-abstractions==1.9.9",
   "dulwich==1.2.5",
   "pyjwt[crypto]==2.13.0"

api/src/backend/api/adapters.py

@@ -1,6 +1,6 @@
+from allauth.account.models import EmailAddress
+from allauth.core.exceptions import ImmediateHttpResponse
 from allauth.socialaccount.adapter import DefaultSocialAccountAdapter
-from django.db import transaction
-
 from api.db_router import MainRouter
 from api.db_utils import rls_transaction
 from api.models import (
@@ -11,6 +11,9 @@ from api.models import (
     User,
     UserRoleRelationship,
 )
+from api.utils import accept_invitation_for_user
+from django.db import transaction
+from django.http import HttpResponseForbidden
 
 
 class ProwlerSocialAccountAdapter(DefaultSocialAccountAdapter):
@@ -21,9 +24,30 @@ class ProwlerSocialAccountAdapter(DefaultSocialAccountAdapter):
         except User.DoesNotExist:
             return None
 
+    @staticmethod
+    def _get_invitation_token(request):
+        for source_name in ("data", "POST"):
+            data = getattr(request, source_name, None) or {}
+            if not hasattr(data, "get"):
+                continue
+            invitation_token = data.get("invitation_token")
+            if invitation_token:
+                return invitation_token
+
+        wrapped_request = getattr(request, "_request", None)
+        if wrapped_request and wrapped_request is not request:
+            return ProwlerSocialAccountAdapter._get_invitation_token(wrapped_request)
+
+        return None
+
     def pre_social_login(self, request, sociallogin):
-        # Link existing accounts with the same email address
-        email = sociallogin.account.extra_data.get("email")
+        # The provider account is already bound, so no email-based linking is needed.
+        if sociallogin.account.pk:
+            return
+
+        # Prefer the normalized email populated by allauth. GitHub can return the
+        # primary email separately from the profile stored in extra_data.
+        email = sociallogin.user.email or sociallogin.account.extra_data.get("email")
         if sociallogin.provider.id == "saml":
             # For SAML, the asserted NameID email cannot be trusted on its own:
             # any tenant can claim any email domain in its SAML configuration. To
@@ -64,6 +88,17 @@ class ProwlerSocialAccountAdapter(DefaultSocialAccountAdapter):
         if email:
             existing_user = self.get_user_by_email(email)
             if existing_user:
+                email_is_verified = EmailAddress.objects.filter(
+                    user=existing_user,
+                    email__iexact=email,
+                    verified=True,
+                ).exists()
+                provider_verified_email = any(
+                    address.verified and address.email.casefold() == email.casefold()
+                    for address in sociallogin.email_addresses
+                )
+                if not email_is_verified or not provider_verified_email:
+                    raise ImmediateHttpResponse(HttpResponseForbidden())
                 sociallogin.connect(request, existing_user)
 
     def save_user(self, request, sociallogin, form=None):
@@ -84,29 +119,38 @@ class ProwlerSocialAccountAdapter(DefaultSocialAccountAdapter):
                     user.name = social_account_name
                     user.save(using=MainRouter.admin_db)
 
-                tenant = Tenant.objects.using(MainRouter.admin_db).create(
-                    name=f"{user.email.split('@')[0]} default tenant"
-                )
-                with rls_transaction(str(tenant.id)):
-                    Membership.objects.using(MainRouter.admin_db).create(
-                        user=user, tenant=tenant, role=Membership.RoleChoices.OWNER
-                    )
-                    role = Role.objects.using(MainRouter.admin_db).create(
-                        name="admin",
-                        tenant_id=tenant.id,
-                        manage_users=True,
-                        manage_account=True,
-                        manage_billing=True,
-                        manage_providers=True,
-                        manage_integrations=True,
-                        manage_scans=True,
-                        unlimited_visibility=True,
-                    )
-                    UserRoleRelationship.objects.using(MainRouter.admin_db).create(
+                invitation_token = self._get_invitation_token(request)
+                if invitation_token:
+                    invitation, _ = accept_invitation_for_user(
                         user=user,
-                        role=role,
-                        tenant_id=tenant.id,
+                        invitation_token=invitation_token,
                     )
+                    request.prowler_invitation_token = invitation_token
+                    request.prowler_invitation_tenant_id = str(invitation.tenant_id)
+                else:
+                    tenant = Tenant.objects.using(MainRouter.admin_db).create(
+                        name=f"{user.email.split('@')[0]} default tenant"
+                    )
+                    with rls_transaction(str(tenant.id)):
+                        Membership.objects.using(MainRouter.admin_db).create(
+                            user=user, tenant=tenant, role=Membership.RoleChoices.OWNER
+                        )
+                        role = Role.objects.using(MainRouter.admin_db).create(
+                            name="admin",
+                            tenant_id=tenant.id,
+                            manage_users=True,
+                            manage_account=True,
+                            manage_billing=True,
+                            manage_providers=True,
+                            manage_integrations=True,
+                            manage_scans=True,
+                            unlimited_visibility=True,
+                        )
+                        UserRoleRelationship.objects.using(MainRouter.admin_db).create(
+                            user=user,
+                            role=role,
+                            tenant_id=tenant.id,
+                        )
             else:
                 request.session["saml_user_created"] = str(user.id)
 

api/src/backend/api/apps.py

@@ -1,14 +1,12 @@
 import logging
 import os
 import sys
-
 from pathlib import Path
 
-from django.apps import AppConfig
-from django.conf import settings
-
 from config.custom_logging import BackendLogger
 from config.env import env
+from django.apps import AppConfig
+from django.conf import settings
 
 logger = logging.getLogger(BackendLogger.API)
 
@@ -30,8 +28,10 @@ class ApiConfig(AppConfig):
     name = "api"
 
     def ready(self):
-        from api import schema_extensions  # noqa: F401
-        from api import signals  # noqa: F401
+        from api import (
+            schema_extensions,  # noqa: F401
+            signals,  # noqa: F401
+        )
 
         # Generate required cryptographic keys if not present, but only if:
         #   `"manage.py" not in sys.argv[0]`: If an external server (e.g., Gunicorn) is running the app
@@ -42,9 +42,6 @@ class ApiConfig(AppConfig):
         ):
             self._ensure_crypto_keys()
 
-        # Neo4j driver is created lazily on first use (see api.attack_paths.database).
-        # App init never contacts Neo4j, so a Neo4j outage cannot block API startup.
-
     def _ensure_crypto_keys(self):
         """
         Orchestrator method that ensures all required cryptographic keys are present.

api/src/backend/api/attack_paths/__init__.py

@@ -5,7 +5,6 @@ from api.attack_paths.queries import (
     get_query_by_id,
 )
 
-
 __all__ = [
     "AttackPathsQueryDefinition",
     "AttackPathsQueryParameterDefinition",

api/src/backend/api/attack_paths/cypher_sanitizer.py

@@ -4,10 +4,10 @@ Cypher sanitizer for custom (user-supplied) Attack Paths queries.
 Two responsibilities:
 
 1. **Validation** - reject queries containing SSRF or dangerous procedure
-   patterns (defense-in-depth; the primary control is ``neo4j.READ_ACCESS``).
+   patterns (defense-in-depth; the primary control is `neo4j.READ_ACCESS`).
 
 2. **Provider-scoped label injection** - inject a dynamic
-   ``_Provider_{uuid}`` label into every node pattern so the database can
+   `_Provider_{uuid}` label into every node pattern so the database can
    use its native label index for provider isolation.
 
 Label-injection pipeline:
@@ -22,18 +22,16 @@ Label-injection pipeline:
 import re
 
 from rest_framework.exceptions import ValidationError
-
 from tasks.jobs.attack_paths.config import get_provider_label
 
-
 # Step 1 - String / comment protection
-# Single combined regex: strings first, then line comments.
+# Single combined regex: strings first, then line comments
 # The regex engine finds the leftmost match, so a string like 'https://prowler.com'
-# is consumed as a string before the // inside it can match as a comment.
+# is consumed as a string before the // inside it can match as a comment
 _PROTECTED_RE = re.compile(r"'(?:[^'\\]|\\.)*'|\"(?:[^\"\\]|\\.)*\"|//[^\n]*")
 
 # Step 2 - Clause splitting
-# OPTIONAL MATCH must come before MATCH to avoid partial matching.
+# `OPTIONAL MATCH` must come before `MATCH` to avoid partial matching
 _CLAUSE_RE = re.compile(
     r"\b(OPTIONAL\s+MATCH|MATCH|WHERE|RETURN|WITH|ORDER\s+BY"
     r"|SKIP|LIMIT|UNION|UNWIND|CALL)\b",
@@ -41,10 +39,10 @@ _CLAUSE_RE = re.compile(
 )
 
 # Pass A - Labeled node patterns (all segments)
-# Matches node patterns that have at least one :Label.
-# (?<!\w)\(  - open paren NOT preceded by a word char (excludes function calls).
-# Group 1:  optional variable + one or more :Label
-# Group 2:  optional {properties} + closing paren
+# Matches node patterns that have at least one `:Label`
+# `(?<!\w)\(`  - open paren NOT preceded by a word char, excludes function calls
+# Group 1:  optional variable + one or more `:Label`
+# Group 2:  optional `{`properties`}` + closing paren
 _LABELED_NODE_RE = re.compile(
     r"(?<!\w)\("
     r"("
@@ -57,9 +55,9 @@ _LABELED_NODE_RE = re.compile(
     r")"
 )
 
-# Pass B - Bare node patterns (MATCH segments only)
-# Matches (identifier) or (identifier {properties}) without any :Label.
-# Only applied in MATCH/OPTIONAL MATCH segments.
+# Pass B - Bare node patterns (`MATCH` segments only)
+# Matches (identifier) or (identifier {properties}) without any `:Label`
+# Only applied in `MATCH` / `OPTIONAL MATCH` segments
 _BARE_NODE_RE = re.compile(
     r"(?<!\w)\(" r"(\s*[a-zA-Z_]\w*)" r"(\s*(?:\{[^}]*\})?)" r"\s*\)"
 )
@@ -98,6 +96,11 @@ def inject_provider_label(cypher: str, provider_id: str) -> str:
         node pattern.
     """
     label = get_provider_label(provider_id)
+    return inject_label(cypher, label)
+
+
+def inject_label(cypher: str, label: str) -> str:
+    """Rewrite a Cypher query to append a label to every node pattern."""
 
     # Step 1: Protect strings and comments (single pass, leftmost-first)
     protected: list[str] = []
@@ -136,9 +139,7 @@ def inject_provider_label(cypher: str, provider_id: str) -> str:
     return work
 
 
-# ---------------------------------------------------------------------------
 # Validation
-# ---------------------------------------------------------------------------
 
 # Patterns that indicate SSRF or dangerous procedure calls
 # Defense-in-depth layer - the primary control is `neo4j.READ_ACCESS`

api/src/backend/api/attack_paths/database.py

@@ -1,263 +1,33 @@
-import atexit
-import logging
-import threading
+"""Backwards-compatible facade over the ingest and sink modules.
 
-from contextlib import contextmanager
-from typing import Any, Iterator
+Historically this module owned a single Neo4j driver used for both the
+cartography temp database and the per-tenant sink database. The port to AWS
+Neptune split those roles: the cartography ingest (temp) database is always
+Neo4j and lives in `api.attack_paths.ingest`; the sink is configurable
+(Neo4j or Neptune) and lives in `api.attack_paths.sink`. This shim preserves
+the public API that `tasks/` and `api/v1/views.py` already depend on, and
+dispatches to the right module by database-name prefix.
+
+A database name starting with `db-tmp-scan-` is a cartography temp DB and
+routes to ingest. Everything else routes to the configured sink.
+"""
+
+from contextlib import AbstractContextManager
+from typing import Any
 from uuid import UUID
 
-import neo4j
-import neo4j.exceptions
-
+import neo4j  # noqa: F401 - kept for tests that patch api.attack_paths.database.neo4j
+from api.attack_paths import ingest
+from api.attack_paths import sink as sink_module
 from config.env import env
-from django.conf import settings
-
-from api.attack_paths.retryable_session import RetryableSession
-from tasks.jobs.attack_paths.config import (
-    BATCH_SIZE,
-    PROVIDER_RESOURCE_LABEL,
-    get_provider_label,
+from django.conf import (
+    settings,  # noqa: F401 - kept for tests that patch ...database.settings
 )
 
-# Without this Celery goes crazy with Neo4j logging
-logging.getLogger("neo4j").setLevel(logging.ERROR)
-logging.getLogger("neo4j").propagate = False
-
-SERVICE_UNAVAILABLE_MAX_RETRIES = env.int(
-    "ATTACK_PATHS_SERVICE_UNAVAILABLE_MAX_RETRIES", default=3
-)
-READ_QUERY_TIMEOUT_SECONDS = env.int(
-    "ATTACK_PATHS_READ_QUERY_TIMEOUT_SECONDS", default=30
-)
 MAX_CUSTOM_QUERY_NODES = env.int("ATTACK_PATHS_MAX_CUSTOM_QUERY_NODES", default=250)
-# Shorter than CONN_ACQUISITION_TIMEOUT — the driver requires acquisition to be
-# the longer of the two (it may include opening a new connection).
-CONNECTION_TIMEOUT = env.int("NEO4J_CONNECTION_TIMEOUT", default=5)
-CONN_ACQUISITION_TIMEOUT = env.int("NEO4J_CONN_ACQUISITION_TIMEOUT", default=15)
-READ_EXCEPTION_CODES = [
-    "Neo.ClientError.Statement.AccessMode",
-    "Neo.ClientError.Procedure.ProcedureNotFound",
-]
-CLIENT_STATEMENT_EXCEPTION_PREFIX = "Neo.ClientError.Statement."
 
-# Module-level process-wide driver singleton
-_driver: neo4j.Driver | None = None
-_lock = threading.Lock()
-
-# Base Neo4j functions
-
-
-def get_uri() -> str:
-    host = settings.DATABASES["neo4j"]["HOST"]
-    port = settings.DATABASES["neo4j"]["PORT"]
-    return f"bolt://{host}:{port}"
-
-
-def init_driver() -> neo4j.Driver:
-    global _driver
-    if _driver is not None:
-        return _driver
-
-    with _lock:
-        if _driver is None:
-            uri = get_uri()
-            config = settings.DATABASES["neo4j"]
-
-            driver = neo4j.GraphDatabase.driver(
-                uri,
-                auth=(config["USER"], config["PASSWORD"]),
-                keep_alive=True,
-                max_connection_lifetime=7200,
-                connection_timeout=CONNECTION_TIMEOUT,
-                connection_acquisition_timeout=CONN_ACQUISITION_TIMEOUT,
-                max_connection_pool_size=50,
-            )
-            # Publish the singleton only after connectivity is verified so a
-            # failed probe does not leave an unverified driver behind. Close the
-            # driver on failure so a repeatedly-probed outage cannot leak pools.
-            try:
-                driver.verify_connectivity()
-            except Exception:
-                driver.close()
-                raise
-            _driver = driver
-
-            # Register cleanup handler (only runs once since we're inside the _driver is None block)
-            atexit.register(close_driver)
-
-    return _driver
-
-
-def get_driver() -> neo4j.Driver:
-    return init_driver()
-
-
-def close_driver() -> None:  # TODO: Use it
-    global _driver
-    with _lock:
-        if _driver is not None:
-            try:
-                _driver.close()
-
-            finally:
-                _driver = None
-
-
-@contextmanager
-def get_session(
-    database: str | None = None, default_access_mode: str | None = None
-) -> Iterator[RetryableSession]:
-    session_wrapper: RetryableSession | None = None
-
-    try:
-        session_wrapper = RetryableSession(
-            session_factory=lambda: get_driver().session(
-                database=database, default_access_mode=default_access_mode
-            ),
-            max_retries=SERVICE_UNAVAILABLE_MAX_RETRIES,
-        )
-        yield session_wrapper
-
-    except neo4j.exceptions.Neo4jError as exc:
-        if (
-            default_access_mode == neo4j.READ_ACCESS
-            and exc.code
-            and exc.code in READ_EXCEPTION_CODES
-        ):
-            message = "Read query not allowed"
-            code = READ_EXCEPTION_CODES[0]
-            raise WriteQueryNotAllowedException(message=message, code=code)
-
-        message = exc.message if exc.message is not None else str(exc)
-
-        if exc.code and exc.code.startswith(CLIENT_STATEMENT_EXCEPTION_PREFIX):
-            raise ClientStatementException(message=message, code=exc.code)
-
-        raise GraphDatabaseQueryException(message=message, code=exc.code)
-
-    finally:
-        if session_wrapper is not None:
-            session_wrapper.close()
-
-
-def execute_read_query(
-    database: str,
-    cypher: str,
-    parameters: dict[str, Any] | None = None,
-) -> neo4j.graph.Graph:
-    with get_session(database, default_access_mode=neo4j.READ_ACCESS) as session:
-
-        def _run(tx: neo4j.ManagedTransaction) -> neo4j.graph.Graph:
-            result = tx.run(
-                cypher, parameters or {}, timeout=READ_QUERY_TIMEOUT_SECONDS
-            )
-            return result.graph()
-
-        return session.execute_read(_run)
-
-
-def create_database(database: str) -> None:
-    query = "CREATE DATABASE $database IF NOT EXISTS"
-    parameters = {"database": database}
-
-    with get_session() as session:
-        session.run(query, parameters)
-
-
-def drop_database(database: str) -> None:
-    query = f"DROP DATABASE `{database}` IF EXISTS DESTROY DATA"
-
-    with get_session() as session:
-        session.run(query)
-
-
-def drop_subgraph(database: str, provider_id: str) -> int:
-    """
-    Delete all nodes for a provider from the tenant database.
-
-    Deletes relationships then nodes in batches (not `DETACH DELETE`) so a dense
-    provider's graph cannot exceed Neo4j's transaction memory limit.
-    Silently returns 0 if the database doesn't exist.
-    """
-    provider_label = get_provider_label(provider_id)
-    deleted_nodes = 0
-
-    try:
-        with get_session(database) as session:
-            # Phase 1: delete relationships incident to provider nodes in batches.
-            deleted_count = 1
-            while deleted_count > 0:
-                result = session.run(
-                    f"""
-                    MATCH (:`{provider_label}`)-[r]-()
-                    WITH DISTINCT r LIMIT $batch_size
-                    DELETE r
-                    RETURN COUNT(r) AS deleted_rels_count
-                    """,
-                    {"batch_size": BATCH_SIZE},
-                )
-                deleted_count = result.single().get("deleted_rels_count", 0)
-
-            # Phase 2: delete the now relationship-free nodes in batches.
-            deleted_count = 1
-            while deleted_count > 0:
-                result = session.run(
-                    f"""
-                    MATCH (n:{PROVIDER_RESOURCE_LABEL}:`{provider_label}`)
-                    WITH n LIMIT $batch_size
-                    DELETE n
-                    RETURN COUNT(n) AS deleted_nodes_count
-                    """,
-                    {"batch_size": BATCH_SIZE},
-                )
-                deleted_count = result.single().get("deleted_nodes_count", 0)
-                deleted_nodes += deleted_count
-
-    except GraphDatabaseQueryException as exc:
-        if exc.code == "Neo.ClientError.Database.DatabaseNotFound":
-            return 0
-        raise
-
-    return deleted_nodes
-
-
-def has_provider_data(database: str, provider_id: str) -> bool:
-    """
-    Check if any ProviderResource node exists for this provider.
-
-    Returns `False` if the database doesn't exist.
-    """
-    provider_label = get_provider_label(provider_id)
-    query = f"MATCH (n:{PROVIDER_RESOURCE_LABEL}:`{provider_label}`) RETURN 1 LIMIT 1"
-
-    try:
-        with get_session(database, default_access_mode=neo4j.READ_ACCESS) as session:
-            result = session.run(query)
-            return result.single() is not None
-
-    except GraphDatabaseQueryException as exc:
-        if exc.code == "Neo.ClientError.Database.DatabaseNotFound":
-            return False
-        raise
-
-
-def clear_cache(database: str) -> None:
-    query = "CALL db.clearQueryCaches()"
-
-    try:
-        with get_session(database) as session:
-            session.run(query)
-
-    except GraphDatabaseQueryException as exc:
-        logging.warning(f"Failed to clear query cache for database `{database}`: {exc}")
-
-
-# Neo4j functions related to Prowler + Cartography
-
-
-def get_database_name(entity_id: str | UUID, temporary: bool = False) -> str:
-    prefix = "tmp-scan" if temporary else "tenant"
-    return f"db-{prefix}-{str(entity_id).lower()}"
+TEMP_DB_PREFIX = "db-tmp-scan-"
+DATABASE_NOT_FOUND_CODE = "Neo.ClientError.Database.DatabaseNotFound"
 
 
 # Exceptions
@@ -272,13 +42,190 @@ class GraphDatabaseQueryException(Exception):
     def __str__(self) -> str:
         if self.code:
             return f"{self.code}: {self.message}"
-
         return self.message
 
 
+class NeptuneWriteRetryExhaustedException(GraphDatabaseQueryException):
+    pass
+
+
 class WriteQueryNotAllowedException(GraphDatabaseQueryException):
     pass
 
 
 class ClientStatementException(GraphDatabaseQueryException):
     pass
+
+
+# Routing
+
+
+def _is_ingest_database(database: str | None) -> bool:
+    return bool(database) and database.startswith(TEMP_DB_PREFIX)
+
+
+# Driver lifecycle
+
+
+def init_driver() -> Any:
+    """Initialize the configured sink backend.
+
+    The ingest driver (Neo4j for cartography temp DBs) stays lazy: it is
+    only initialized when a temp-DB operation actually runs, which never
+    happens on API pods.
+    """
+    return sink_module.init()
+
+
+def close_driver() -> None:
+    """Close every driver held by this process."""
+    sink_module.close()
+    ingest.close_driver()
+
+
+def get_driver() -> neo4j.Driver:
+    """Return the sink backend's underlying driver.
+
+    Only meaningful for the Neo4j sink (where the backend has a single Neo4j
+    driver). On Neptune this returns the writer driver. Kept for tests and
+    legacy call-sites; prefer `get_session` for new code.
+    """
+    backend = sink_module.get_backend()
+
+    # Neo4jSink exposes get_driver(); NeptuneSink exposes get_writer()
+    if hasattr(backend, "get_driver"):
+        return backend.get_driver()
+
+    if hasattr(backend, "get_writer"):
+        return backend.get_writer()
+
+    raise RuntimeError("Active sink backend does not expose a driver handle")
+
+
+def verify_connectivity() -> None:
+    """Raise if the configured graph database is unreachable on the API read path.
+
+    Backend-agnostic entry point for the readiness probe: Neo4j verifies its
+    driver, Neptune verifies the reader endpoint.
+    """
+    sink_module.get_backend().verify_connectivity()
+
+
+def verify_scan_databases_available() -> None:
+    """Raise if either graph database needed by an Attack Paths scan is unavailable."""
+    errors: list[str] = []
+    first_error: Exception | None = None
+
+    try:
+        ingest.get_driver().verify_connectivity()
+    except Exception as exc:
+        errors.append(f"ingest Neo4j: {exc}")
+        first_error = exc
+
+    try:
+        get_driver().verify_connectivity()
+    except Exception as exc:
+        errors.append(f"sink {settings.ATTACK_PATHS_SINK_DATABASE}: {exc}")
+        if first_error is None:
+            first_error = exc
+
+    if errors:
+        raise RuntimeError(
+            "Attack Paths graph database unavailable before scan start: "
+            + "; ".join(errors)
+        ) from first_error
+
+
+def get_uri() -> str:
+    """Return the sink URI. Retained for backwards compatibility."""
+    if settings.ATTACK_PATHS_SINK_DATABASE == "neptune":
+        cfg = settings.DATABASES["neptune"]
+        return f"bolt+s://{cfg['WRITER_ENDPOINT']}:{cfg['PORT']}"
+
+    cfg = settings.DATABASES["neo4j"]
+    return f"bolt://{cfg['HOST']}:{cfg['PORT']}"
+
+
+def get_ingest_uri() -> str:
+    """Neo4j URI for the cartography temp (ingest) database, which is always
+    Neo4j regardless of the configured sink."""
+    return ingest.get_uri()
+
+
+# Session API
+
+
+def get_session(
+    database: str | None = None,
+    default_access_mode: str | None = None,
+) -> AbstractContextManager:
+    """Return a session against the right backend.
+
+    - `database` names starting with `db-tmp-scan-` always go to ingest.
+    - No database name → ingest (used for CREATE / DROP DATABASE admin ops).
+    - Any other name → sink.
+    """
+    if _is_ingest_database(database) or database is None:
+        return ingest.get_session(
+            database=database, default_access_mode=default_access_mode
+        )
+
+    return sink_module.get_backend().get_session(
+        database=database, default_access_mode=default_access_mode
+    )
+
+
+def execute_read_query(
+    database: str,
+    cypher: str,
+    parameters: dict[str, Any] | None = None,
+) -> neo4j.graph.Graph:
+    """Read-only query against the sink."""
+    return sink_module.get_backend().execute_read_query(database, cypher, parameters)
+
+
+def create_database(database: str) -> None:
+    """Create a database. Temp DBs always land on ingest (Neo4j).
+
+    On the Neo4j sink, tenant DBs also route to ingest because both drivers
+    connect to the same Neo4j cluster. On the Neptune sink, tenant DB creates
+    are no-ops.
+    """
+    if _is_ingest_database(database):
+        ingest.create_database(database)
+        return
+
+    sink_module.get_backend().create_database(database)
+
+
+def drop_database(database: str) -> None:
+    """Drop a database. Mirrors `create_database` routing."""
+    if _is_ingest_database(database):
+        ingest.drop_database(database)
+        return
+
+    sink_module.get_backend().drop_database(database)
+
+
+def drop_subgraph(database: str, provider_id: str) -> int:
+    return sink_module.get_backend().drop_subgraph(database, provider_id)
+
+
+def has_provider_data(database: str, provider_id: str) -> bool:
+    return sink_module.get_backend().has_provider_data(database, provider_id)
+
+
+def clear_cache(database: str) -> None:
+    if _is_ingest_database(database):
+        ingest.clear_cache(database)
+        return
+
+    sink_module.get_backend().clear_cache(database)
+
+
+# Name helper
+
+
+def get_database_name(entity_id: str | UUID, temporary: bool = False) -> str:
+    prefix = "tmp-scan" if temporary else "tenant"
+    return f"db-{prefix}-{str(entity_id).lower()}"

api/src/backend/api/attack_paths/ingest/__init__.py

@@ -0,0 +1,29 @@
+"""Cartography ingest layer.
+
+Public surface for the per-scan Neo4j temp database driver. Implementation
+lives in `api.attack_paths.ingest.driver`.
+"""
+
+from api.attack_paths.ingest.driver import (
+    clear_cache,
+    close_driver,
+    create_database,
+    drop_database,
+    get_driver,
+    get_session,
+    get_uri,
+    init_driver,
+    run_cypher,
+)
+
+__all__ = [
+    "clear_cache",
+    "close_driver",
+    "create_database",
+    "drop_database",
+    "get_driver",
+    "get_session",
+    "get_uri",
+    "init_driver",
+    "run_cypher",
+]

api/src/backend/api/attack_paths/ingest/driver.py

@@ -0,0 +1,187 @@
+"""Cartography ingest driver: per-scan throw-away Neo4j database.
+
+Cartography writes each scan's graph into a throw-away Neo4j database named
+`db-tmp-scan-{scan_uuid}`. This is always Neo4j, regardless of the configured
+sink: Neptune is single-database and cannot host per-scan throw-away
+databases. This module owns the Neo4j driver used for those temp DBs and the
+admin ops they need (CREATE / DROP DATABASE).
+"""
+
+import atexit
+import logging
+import threading
+from collections.abc import Iterator
+from contextlib import contextmanager
+from typing import Any
+
+import neo4j
+import neo4j.exceptions
+from api.attack_paths.retryable_session import RetryableSession
+from config.env import env
+from django.conf import settings
+
+logging.getLogger("neo4j").setLevel(logging.ERROR)
+logging.getLogger("neo4j").propagate = False
+
+SERVICE_UNAVAILABLE_MAX_RETRIES = env.int(
+    "ATTACK_PATHS_SERVICE_UNAVAILABLE_MAX_RETRIES", default=3
+)
+CONN_ACQUISITION_TIMEOUT = env.int("NEO4J_CONN_ACQUISITION_TIMEOUT", default=15)
+# TCP connect timeout, ordered below the acquisition timeout so an unreachable
+# host can't pin a worker on a temp-DB op longer than this.
+CONNECTION_TIMEOUT = env.int("NEO4J_CONNECTION_TIMEOUT", default=5)
+MAX_CONNECTION_LIFETIME = env.int("NEO4J_MAX_CONNECTION_LIFETIME", default=7200)
+MAX_CONNECTION_POOL_SIZE = env.int("NEO4J_MAX_CONNECTION_POOL_SIZE", default=50)
+
+_driver: neo4j.Driver | None = None
+_lock = threading.Lock()
+
+
+def _neo4j_config() -> dict:
+    return settings.DATABASES["neo4j"]
+
+
+def get_uri() -> str:
+    """Bolt URI for the Neo4j temp (ingest) database. Always Neo4j."""
+    config = _neo4j_config()
+    host = config["HOST"]
+    port = config["PORT"]
+    if not host or not port:
+        raise RuntimeError(
+            "NEO4J_HOST / NEO4J_PORT must be set to use the attack-paths "
+            "temp database. Workers require Neo4j env even when the sink is Neptune."
+        )
+
+    return f"bolt://{host}:{port}"
+
+
+def init_driver() -> neo4j.Driver:
+    """Initialize the temp-database Neo4j driver. Idempotent."""
+    global _driver
+    if _driver is not None:
+        return _driver
+
+    with _lock:
+        if _driver is None:
+            config = _neo4j_config()
+            _driver = neo4j.GraphDatabase.driver(
+                get_uri(),
+                auth=(config["USER"], config["PASSWORD"]),
+                keep_alive=True,
+                max_connection_lifetime=MAX_CONNECTION_LIFETIME,
+                connection_timeout=CONNECTION_TIMEOUT,
+                connection_acquisition_timeout=CONN_ACQUISITION_TIMEOUT,
+                max_connection_pool_size=MAX_CONNECTION_POOL_SIZE,
+            )
+            # Best-effort connectivity check: a Neo4j that is down at boot must
+            # not crash the worker. The driver reconnects lazily on first use.
+            try:
+                _driver.verify_connectivity()
+
+            except Exception:
+                logging.warning(
+                    "Neo4j temp-database unreachable at init; continuing with a "
+                    "lazily-reconnecting driver",
+                    exc_info=True,
+                )
+
+            atexit.register(close_driver)
+
+    return _driver
+
+
+def get_driver() -> neo4j.Driver:
+    return init_driver()
+
+
+def close_driver() -> None:
+    global _driver
+    with _lock:
+        if _driver is not None:
+            try:
+                _driver.close()
+            finally:
+                _driver = None
+
+
+@contextmanager
+def get_session(
+    database: str | None = None,
+    default_access_mode: str | None = None,
+) -> Iterator[RetryableSession]:
+    """Session against the Neo4j temp-database cluster. Used for temp DB sessions
+    and for admin operations (CREATE / DROP DATABASE) when `database` is None."""
+    from api.attack_paths.database import (
+        ClientStatementException,
+        GraphDatabaseQueryException,
+        WriteQueryNotAllowedException,
+    )
+
+    READ_EXCEPTION_CODES = [
+        "Neo.ClientError.Statement.AccessMode",
+        "Neo.ClientError.Procedure.ProcedureNotFound",
+    ]
+    CLIENT_STATEMENT_EXCEPTION_PREFIX = "Neo.ClientError.Statement."
+
+    session_wrapper: RetryableSession | None = None
+    try:
+        session_wrapper = RetryableSession(
+            session_factory=lambda: get_driver().session(
+                database=database, default_access_mode=default_access_mode
+            ),
+            max_retries=SERVICE_UNAVAILABLE_MAX_RETRIES,
+        )
+        yield session_wrapper
+
+    except neo4j.exceptions.Neo4jError as exc:
+        if (
+            default_access_mode == neo4j.READ_ACCESS
+            and exc.code
+            and exc.code in READ_EXCEPTION_CODES
+        ):
+            raise WriteQueryNotAllowedException(
+                message="Read query not allowed", code=READ_EXCEPTION_CODES[0]
+            )
+
+        message = exc.message if exc.message is not None else str(exc)
+        if exc.code and exc.code.startswith(CLIENT_STATEMENT_EXCEPTION_PREFIX):
+            raise ClientStatementException(message=message, code=exc.code)
+        raise GraphDatabaseQueryException(message=message, code=exc.code)
+
+    finally:
+        if session_wrapper is not None:
+            session_wrapper.close()
+
+
+def create_database(database: str) -> None:
+    """Create a database on the Neo4j cluster. Used for temp scan DBs."""
+    with get_session() as session:
+        session.run("CREATE DATABASE $database IF NOT EXISTS", {"database": database})
+
+
+def drop_database(database: str) -> None:
+    """Drop a database on the Neo4j cluster. Used for temp scan DBs."""
+    with get_session() as session:
+        session.run(f"DROP DATABASE `{database}` IF EXISTS DESTROY DATA")
+
+
+def clear_cache(database: str) -> None:
+    """Best-effort cache clear for a Neo4j database."""
+    from api.attack_paths.database import GraphDatabaseQueryException
+
+    try:
+        with get_session(database) as session:
+            session.run("CALL db.clearQueryCaches()")
+
+    except GraphDatabaseQueryException as exc:
+        logging.warning(f"Failed to clear query cache for database `{database}`: {exc}")
+
+
+def run_cypher(
+    database: str | None,
+    cypher: str,
+    parameters: dict[str, Any] | None = None,
+) -> Any:
+    """Execute Cypher directly without the context manager. Thin helper."""
+    with get_session(database) as session:
+        return session.run(cypher, parameters or {})

api/src/backend/api/attack_paths/queries/__init__.py

@@ -1,12 +1,11 @@
-from api.attack_paths.queries.types import (
-    AttackPathsQueryDefinition,
-    AttackPathsQueryParameterDefinition,
-)
 from api.attack_paths.queries.registry import (
     get_queries_for_provider,
     get_query_by_id,
 )
-
+from api.attack_paths.queries.types import (
+    AttackPathsQueryDefinition,
+    AttackPathsQueryParameterDefinition,
+)
 
 __all__ = [
     "AttackPathsQueryDefinition",

api/src/backend/api/attack_paths/queries/registry.py

@@ -1,13 +1,14 @@
-from api.attack_paths.queries.types import AttackPathsQueryDefinition
 from api.attack_paths.queries.aws import AWS_QUERIES
 
+# TODO: drop after Neptune cutover
+from api.attack_paths.queries.aws_deprecated import AWS_DEPRECATED_QUERIES
+from api.attack_paths.queries.types import AttackPathsQueryDefinition
 
-# Query definitions organized by provider
+# Query definitions for scans synced with the current schema.
 _QUERY_DEFINITIONS: dict[str, list[AttackPathsQueryDefinition]] = {
     "aws": AWS_QUERIES,
 }
 
-# Flat lookup by query ID for O(1) access
 _QUERIES_BY_ID: dict[str, AttackPathsQueryDefinition] = {
     definition.id: definition
     for definitions in _QUERY_DEFINITIONS.values()
@@ -15,11 +16,45 @@ _QUERIES_BY_ID: dict[str, AttackPathsQueryDefinition] = {
 }
 
 
-def get_queries_for_provider(provider: str) -> list[AttackPathsQueryDefinition]:
-    """Get all attack path queries for a specific provider."""
-    return _QUERY_DEFINITIONS.get(provider, [])
+# TODO: drop after Neptune cutover
+#
+# Query definitions for pre-cutover scans (`AttackPathsScan.is_migrated=False`)
+# whose graph data was written under the previous schema. Both maps expose the
+# same query IDs so the API contract is identical regardless of which set is
+# routed to.
+_DEPRECATED_QUERY_DEFINITIONS: dict[str, list[AttackPathsQueryDefinition]] = {
+    "aws": AWS_DEPRECATED_QUERIES,
+}
+
+_DEPRECATED_QUERIES_BY_ID: dict[str, AttackPathsQueryDefinition] = {
+    definition.id: definition
+    for definitions in _DEPRECATED_QUERY_DEFINITIONS.values()
+    for definition in definitions
+}
 
 
-def get_query_by_id(query_id: str) -> AttackPathsQueryDefinition | None:
-    """Get a specific attack path query by its ID."""
-    return _QUERIES_BY_ID.get(query_id)
+def get_queries_for_provider(
+    provider: str,
+    is_migrated: bool = True,
+) -> list[AttackPathsQueryDefinition]:
+    """Get all attack path queries for a provider.
+
+    `is_migrated` selects the catalog: True for scans synced with the current
+    schema, False for pre-cutover scans still using the legacy graph shape.
+    # TODO: drop the `is_migrated` parameter after Neptune cutover
+    """
+    catalog = _QUERY_DEFINITIONS if is_migrated else _DEPRECATED_QUERY_DEFINITIONS
+    return catalog.get(provider, [])
+
+
+def get_query_by_id(
+    query_id: str,
+    is_migrated: bool = True,
+) -> AttackPathsQueryDefinition | None:
+    """Get a specific attack path query by ID.
+
+    `is_migrated` selects the catalog (see `get_queries_for_provider`).
+    # TODO: drop the `is_migrated` parameter after Neptune cutover
+    """
+    by_id = _QUERIES_BY_ID if is_migrated else _DEPRECATED_QUERIES_BY_ID
+    return by_id.get(query_id)

api/src/backend/api/attack_paths/retryable_session.py

@@ -1,5 +1,6 @@
 import logging
-
+import random
+import time
 from collections.abc import Callable
 from typing import Any
 
@@ -9,18 +10,44 @@ import neo4j.exceptions
 logger = logging.getLogger(__name__)
 
 
+class RetryExhaustedError(Exception):
+    def __init__(
+        self,
+        *,
+        retry_context: str,
+        method_name: str,
+        attempts: int,
+        elapsed_seconds: float,
+        last_error: Exception,
+    ) -> None:
+        self.retry_context = retry_context
+        self.method_name = method_name
+        self.attempts = attempts
+        self.elapsed_seconds = elapsed_seconds
+        self.last_error = last_error
+        last_message = getattr(last_error, "message", None) or str(last_error)
+        super().__init__(
+            f"{retry_context} {method_name} failed after {attempts} attempts over "
+            f"{elapsed_seconds:.3f}s. Last error: {last_message}"
+        )
+
+
 class RetryableSession:
-    """
-    Wrapper around `neo4j.Session` that retries `neo4j.exceptions.ServiceUnavailable` errors.
-    """
+    """Wrapper around ``neo4j.Session`` with a refreshable retry policy."""
 
     def __init__(
         self,
         session_factory: Callable[[], neo4j.Session],
         max_retries: int,
+        retry_if: Callable[[Exception], bool] | None = None,
+        initial_retry_delay_seconds: float = 0,
+        retry_context: str | None = None,
     ) -> None:
         self._session_factory = session_factory
         self._max_retries = max(0, max_retries)
+        self._retry_if = retry_if
+        self._initial_retry_delay_seconds = max(0.0, initial_retry_delay_seconds)
+        self._retry_context = retry_context
         self._session = self._session_factory()
 
     def close(self) -> None:
@@ -51,30 +78,75 @@ class RetryableSession:
     def _call_with_retry(self, method_name: str, *args: Any, **kwargs: Any) -> Any:
         attempt = 0
         last_exc: Exception | None = None
+        started_at = time.monotonic()
 
         while attempt <= self._max_retries:
             try:
                 method = getattr(self._session, method_name)
                 return method(*args, **kwargs)
 
-            except (
-                BrokenPipeError,
-                ConnectionResetError,
-                neo4j.exceptions.ServiceUnavailable,
-            ) as exc:  # pragma: no cover - depends on infra
+            except Exception as exc:
+                if not self._should_retry(exc):
+                    raise
+
                 last_exc = exc
                 attempt += 1
 
                 if attempt > self._max_retries:
+                    if self._retry_context is not None:
+                        raise RetryExhaustedError(
+                            retry_context=self._retry_context,
+                            method_name=method_name,
+                            attempts=attempt,
+                            elapsed_seconds=time.monotonic() - started_at,
+                            last_error=exc,
+                        ) from exc
                     raise
 
-                logger.warning(
-                    f"Neo4j session {method_name} failed with {type(exc).__name__} ({attempt}/{self._max_retries} attempts). Retrying..."
-                )
+                delay = self._retry_delay(attempt)
+                if self._retry_context is not None:
+                    error_message = getattr(exc, "message", None) or str(exc)
+                    logger.warning(
+                        "%s %s failed with %s: %s; retry %s/%s in %.3fs",
+                        self._retry_context,
+                        method_name,
+                        type(exc).__name__,
+                        error_message,
+                        attempt,
+                        self._max_retries,
+                        delay,
+                    )
+                else:
+                    logger.warning(
+                        "Graph session %s failed with %s; retry %s/%s in %.3fs",
+                        method_name,
+                        type(exc).__name__,
+                        attempt,
+                        self._max_retries,
+                        delay,
+                    )
                 self._refresh_session()
+                if delay:
+                    time.sleep(delay)
 
         raise last_exc if last_exc else RuntimeError("Unexpected retry loop exit")
 
+    def _should_retry(self, exc: Exception) -> bool:
+        if isinstance(
+            exc,
+            (
+                BrokenPipeError,
+                ConnectionResetError,
+                neo4j.exceptions.ServiceUnavailable,
+            ),
+        ):
+            return True
+        return self._retry_if(exc) if self._retry_if else False
+
+    def _retry_delay(self, attempt: int) -> float:
+        max_delay = self._initial_retry_delay_seconds * (2**attempt)
+        return random.uniform(max_delay / 2, max_delay) if max_delay else 0
+
     def _refresh_session(self) -> None:
         if self._session is not None:
             try:

api/src/backend/api/attack_paths/sink/__init__.py

@@ -0,0 +1,28 @@
+"""Attack-paths sink database layer.
+
+The sink is the persistent store where attack-paths graphs live after a scan
+finishes. Currently selectable between Neo4j (OSS / local dev default) and
+AWS Neptune (hosted dev/staging/prod). Backend is picked by the
+`ATTACK_PATHS_SINK_DATABASE` setting at process init.
+
+This package exposes the public factory API; the implementation lives in
+`api.attack_paths.sink.factory`.
+"""
+
+from api.attack_paths.sink.factory import (
+    SinkBackend,
+    close,
+    get_backend,
+    get_backend_for_name,
+    get_backend_for_scan,
+    init,
+)
+
+__all__ = [
+    "SinkBackend",
+    "close",
+    "get_backend",
+    "get_backend_for_name",
+    "get_backend_for_scan",
+    "init",
+]

api/src/backend/api/attack_paths/sink/base.py

@@ -0,0 +1,94 @@
+"""Protocol every sink backend must implement."""
+
+from contextlib import AbstractContextManager
+from typing import Any, Protocol
+
+import neo4j
+
+
+class SinkDatabase(Protocol):
+    """Contract for the persistent attack-paths graph store.
+
+    The `database` argument is an opaque identifier passed through from the
+    legacy `database.py` API surface. On Neo4j it is the per-tenant database
+    name (e.g. `db-tenant-{uuid}`). On Neptune it is ignored (the cluster
+    has a single graph, and isolation is label-based).
+    """
+
+    sync_batch_size: int
+
+    def init(self) -> None: ...
+
+    def close(self) -> None: ...
+
+    def verify_connectivity(self) -> None:
+        """Raise if the backend the API read path uses is unreachable.
+
+        Neo4j verifies its single driver. Neptune verifies the reader
+        driver (the endpoint the API serves reads from); on single-endpoint
+        clusters the reader aliases the writer, so that path is covered too.
+        Used by the readiness probe; must not block longer than the caller's
+        probe budget.
+        """
+        ...
+
+    def get_session(
+        self,
+        database: str | None = None,
+        default_access_mode: str | None = None,
+    ) -> AbstractContextManager: ...
+
+    def execute_read_query(
+        self,
+        database: str,
+        cypher: str,
+        parameters: dict[str, Any] | None = None,
+    ) -> neo4j.graph.Graph: ...
+
+    def create_database(self, database: str) -> None: ...
+
+    def drop_database(self, database: str) -> None: ...
+
+    def drop_subgraph(self, database: str, provider_id: str) -> int: ...
+
+    def has_provider_data(self, database: str, provider_id: str) -> bool: ...
+
+    def clear_cache(self, database: str) -> None: ...
+
+    def ensure_sync_indexes(self, database: str) -> None:
+        """Create any index needed for the sync write path.
+
+        Called once at the start of each provider sync; must be idempotent.
+        Neo4j creates a `_provider_element_id` index on `_ProviderResource`;
+        Neptune is a no-op (its `~id` lookup needs no index).
+        """
+        ...
+
+    def write_nodes(
+        self,
+        database: str,
+        labels: str,
+        rows: list[dict[str, Any]],
+    ) -> None:
+        """Upsert a batch of nodes into the sink.
+
+        `labels` is a pre-rendered Cypher label string ready to drop after
+        the node variable (e.g. `` `AWSUser`:`_ProviderResource`:`_Tenant_x` ``).
+        Each row carries `provider_element_id` and `props`.
+        """
+        ...
+
+    def write_relationships(
+        self,
+        database: str,
+        rel_type: str,
+        provider_id: str,
+        rows: list[dict[str, Any]],
+    ) -> None:
+        """Upsert a batch of relationships into the sink.
+
+        Each row carries `start_element_id`, `end_element_id`,
+        `provider_element_id` and `props`. `rel_type` is the relationship
+        type (already a valid Cypher identifier).
+        """
+        ...

api/src/backend/api/attack_paths/sink/drop.py

@@ -0,0 +1,81 @@
+"""Shared batched deletion helpers for sink backends."""
+
+import logging
+import time
+from typing import Any
+
+RELATIONSHIP_DELETE_QUERY_TEMPLATES = {
+    "outgoing relationship": """
+        MATCH (n:`{provider_label}`)-[r]->()
+        WITH r LIMIT $batch_size
+        DELETE r
+        RETURN COUNT(r) AS deleted_rels_count
+        """,
+    "incoming relationship": """
+        MATCH (n:`{provider_label}`)<-[r]-()
+        WITH r LIMIT $batch_size
+        DELETE r
+        RETURN COUNT(r) AS deleted_rels_count
+        """,
+}
+
+NODE_DELETE_QUERY_TEMPLATE = """
+    MATCH (n:{provider_resource_label}:`{provider_label}`)
+    WITH n LIMIT $batch_size
+    DELETE n
+    RETURN COUNT(n) AS deleted_nodes_count
+    """
+
+
+def delete_batches(
+    *,
+    session: Any,
+    logger: logging.Logger,
+    log_target: str,
+    provider_id: str,
+    query: str,
+    phase: str,
+    count_key: str,
+    total_key: str,
+    deleted_key: str,
+    initial_total: int,
+    batch_size: int,
+    drop_t0: float,
+) -> tuple[int, int]:
+    def delete_batch(tx: Any) -> int:
+        record = tx.run(query, {"batch_size": batch_size}).single()
+        return (record[count_key] if record else 0) or 0
+
+    deleted_total = initial_total
+    batches = 0
+    while True:
+        logger.info(
+            "Deleting %s batch from %s "
+            "(provider=%s, batch=%s, total_%s=%s, elapsed=%.3fs)",
+            phase,
+            log_target,
+            provider_id,
+            batches + 1,
+            total_key,
+            deleted_total,
+            time.perf_counter() - drop_t0,
+        )
+        deleted = session.execute_write(delete_batch)
+        if deleted == 0:
+            return deleted_total, batches
+
+        batches += 1
+        deleted_total += deleted
+        logger.info(
+            "Deleted %s batch from %s "
+            "(provider=%s, batch=%s, %s=%s, total_%s=%s, elapsed=%.3fs)",
+            phase,
+            log_target,
+            provider_id,
+            batches,
+            deleted_key,
+            deleted,
+            total_key,
+            deleted_total,
+            time.perf_counter() - drop_t0,
+        )

api/src/backend/api/attack_paths/sink/factory.py

@@ -0,0 +1,134 @@
+"""Sink backend factory and process-wide handle cache.
+
+Picks the active backend from `settings.ATTACK_PATHS_SINK_DATABASE` at first
+use, holds the active backend plus any secondary backends needed to serve
+scans written under the previous configuration, and tears them all down on
+process shutdown. Imported via `from api.attack_paths import sink as
+sink_module`.
+"""
+
+import threading
+from enum import StrEnum, auto
+
+from api.attack_paths.sink.base import SinkDatabase
+from api.models import AttackPathsScan
+from django.conf import settings
+
+# Backend names
+
+
+class SinkBackend(StrEnum):
+    NEO4J = auto()
+    NEPTUNE = auto()
+
+
+# Backend cache
+
+_backend: SinkDatabase | None = None
+_secondary_backends: dict[SinkBackend, SinkDatabase] = {}
+_lock = threading.Lock()
+
+
+def _resolve_setting() -> SinkBackend:
+    raw = settings.ATTACK_PATHS_SINK_DATABASE.lower()
+    try:
+        return SinkBackend(raw)
+
+    except ValueError:
+        valid = sorted(b.value for b in SinkBackend)
+        raise RuntimeError(
+            f"ATTACK_PATHS_SINK_DATABASE must be one of {valid}; got {raw!r}"
+        )
+
+
+def _build_backend(name: SinkBackend) -> SinkDatabase:
+    if name is SinkBackend.NEO4J:
+        from api.attack_paths.sink.neo4j import Neo4jSink
+
+        return Neo4jSink()
+
+    if name is SinkBackend.NEPTUNE:
+        from api.attack_paths.sink.neptune import NeptuneSink
+
+        return NeptuneSink()
+
+    raise RuntimeError(f"Unknown sink backend {name!r}")
+
+
+# Lifecycle
+
+
+def init(name: SinkBackend | str | None = None) -> SinkDatabase:
+    """Initialize the configured sink backend. Idempotent."""
+    global _backend
+    if _backend is not None:
+        return _backend
+
+    with _lock:
+        if _backend is None:
+            resolved = SinkBackend(name) if name else _resolve_setting()
+            backend = _build_backend(resolved)
+            backend.init()
+            _backend = backend
+
+    return _backend
+
+
+def close() -> None:
+    """Close the active backend and every cached secondary backend."""
+    global _backend
+    with _lock:
+        backends = [
+            b for b in (_backend, *_secondary_backends.values()) if b is not None
+        ]
+        _backend = None
+        _secondary_backends.clear()
+
+    for backend in backends:
+        try:
+            backend.close()
+
+        except Exception:  # pragma: no cover - best-effort
+            pass
+
+
+def get_backend() -> SinkDatabase:
+    """Return the active sink. Initializes on first call."""
+    return init()
+
+
+# Per-scan routing
+
+
+def get_backend_for_scan(scan: AttackPathsScan) -> SinkDatabase:
+    """Route reads by the sink that stores this scan's graph."""
+    raw_backend = getattr(scan, "sink_backend", SinkBackend.NEO4J.value)
+    if not isinstance(raw_backend, str):
+        raw_backend = SinkBackend.NEO4J.value
+    return get_backend_for_name(raw_backend)
+
+
+def get_backend_for_name(name: SinkBackend | str) -> SinkDatabase:
+    """Return the backend named by persisted scan metadata."""
+    resolved = SinkBackend(name)
+    if resolved is _resolve_setting():
+        return get_backend()
+
+    return _build_backend_cached(resolved)
+
+
+def _build_backend_cached(name: SinkBackend) -> SinkDatabase:
+    # TODO: drop after Neptune cutover
+    # Needed only during cutover to serve Neo4j-written scans from a Neptune-
+    # configured API pod (and vice versa). Once every scan is on Neptune,
+    # `get_backend_for_scan` becomes a one-liner returning `get_backend()`.
+    if name in _secondary_backends:
+        return _secondary_backends[name]
+
+    with _lock:
+        if name not in _secondary_backends:
+            backend = _build_backend(name)
+            backend.init()
+            _secondary_backends[name] = backend
+
+    return _secondary_backends[name]

api/src/backend/api/attack_paths/sink/neo4j.py

@@ -0,0 +1,419 @@
+"""Neo4j sink implementation.
+
+Owns a Neo4j driver independent from the staging driver. On OSS and local dev
+this is the only sink; on hosted deployments it runs only as a legacy read
+path while phase-1 drains tenant DBs.
+"""
+
+import atexit
+import logging
+import threading
+import time
+from collections.abc import Iterator
+from contextlib import AbstractContextManager, contextmanager
+from typing import Any
+
+import neo4j
+import neo4j.exceptions
+from api.attack_paths.retryable_session import RetryableSession
+from api.attack_paths.sink.base import SinkDatabase
+from api.attack_paths.sink.drop import (
+    NODE_DELETE_QUERY_TEMPLATE,
+    RELATIONSHIP_DELETE_QUERY_TEMPLATES,
+    delete_batches,
+)
+from config.env import env
+from django.conf import settings
+
+logging.getLogger("neo4j").setLevel(logging.ERROR)
+logging.getLogger("neo4j").propagate = False
+
+logger = logging.getLogger(__name__)
+
+SERVICE_UNAVAILABLE_MAX_RETRIES = env.int(
+    "ATTACK_PATHS_SERVICE_UNAVAILABLE_MAX_RETRIES", default=3
+)
+READ_QUERY_TIMEOUT_SECONDS = env.int(
+    "ATTACK_PATHS_READ_QUERY_TIMEOUT_SECONDS", default=30
+)
+CONN_ACQUISITION_TIMEOUT = env.int("NEO4J_CONN_ACQUISITION_TIMEOUT", default=15)
+# TCP connect timeout, ordered below the acquisition timeout so an unreachable
+# host can't pin a request or the readiness probe longer than this.
+CONNECTION_TIMEOUT = env.int("NEO4J_CONNECTION_TIMEOUT", default=5)
+MAX_CONNECTION_LIFETIME = env.int("NEO4J_MAX_CONNECTION_LIFETIME", default=7200)
+MAX_CONNECTION_POOL_SIZE = env.int("NEO4J_MAX_CONNECTION_POOL_SIZE", default=50)
+
+READ_EXCEPTION_CODES = [
+    "Neo.ClientError.Statement.AccessMode",
+    "Neo.ClientError.Procedure.ProcedureNotFound",
+]
+CLIENT_STATEMENT_EXCEPTION_PREFIX = "Neo.ClientError.Statement."
+DATABASE_NOT_FOUND_CODE = "Neo.ClientError.Database.DatabaseNotFound"
+
+
+class Neo4jSink(SinkDatabase):
+    """Neo4j-backed sink. Multi-database cluster; tenant isolation is physical."""
+
+    sync_batch_size = env.int("ATTACK_PATHS_NEO4J_SYNC_BATCH_SIZE", default=1000)
+
+    def __init__(self) -> None:
+        self._driver: neo4j.Driver | None = None
+        self._lock = threading.Lock()
+        self._atexit_registered = False
+
+    # Driver
+
+    def _config(self) -> dict:
+        return settings.DATABASES["neo4j"]
+
+    def _uri(self) -> str:
+        cfg = self._config()
+        host = cfg["HOST"]
+        port = cfg["PORT"]
+        if not host or not port:
+            raise RuntimeError(
+                "NEO4J_HOST / NEO4J_PORT must be set when ATTACK_PATHS_SINK_DATABASE=neo4j"
+            )
+        return f"bolt://{host}:{port}"
+
+    def init(self) -> neo4j.Driver:
+        if self._driver is not None:
+            return self._driver
+        with self._lock:
+            if self._driver is None:
+                cfg = self._config()
+                self._driver = neo4j.GraphDatabase.driver(
+                    self._uri(),
+                    auth=(cfg["USER"], cfg["PASSWORD"]),
+                    keep_alive=True,
+                    max_connection_lifetime=MAX_CONNECTION_LIFETIME,
+                    connection_timeout=CONNECTION_TIMEOUT,
+                    connection_acquisition_timeout=CONN_ACQUISITION_TIMEOUT,
+                    max_connection_pool_size=MAX_CONNECTION_POOL_SIZE,
+                )
+                # Eager connectivity check is best-effort:
+                # A Neo4j that is down at boot must not crash the process, same degradation model as Postgres
+                # The driver reconnects lazily on first use
+                # /health/ready surfaces the outage until it recovers
+                try:
+                    self._driver.verify_connectivity()
+
+                except Exception:
+                    logger.warning(
+                        "Neo4j sink unreachable at init; continuing with a lazily-reconnecting driver",
+                        exc_info=True,
+                    )
+
+                if not self._atexit_registered:
+                    atexit.register(self.close)
+                    self._atexit_registered = True
+        return self._driver
+
+    def _get_driver(self) -> neo4j.Driver:
+        return self.init()
+
+    def verify_connectivity(self) -> None:
+        self._get_driver().verify_connectivity()
+
+    def close(self) -> None:
+        with self._lock:
+            if self._driver is not None:
+                try:
+                    self._driver.close()
+                finally:
+                    self._driver = None
+
+    # Sessions
+
+    @contextmanager
+    def get_session(
+        self,
+        database: str | None = None,
+        default_access_mode: str | None = None,
+    ) -> Iterator[RetryableSession]:
+        from api.attack_paths.database import (
+            ClientStatementException,
+            GraphDatabaseQueryException,
+            WriteQueryNotAllowedException,
+        )
+
+        session_wrapper: RetryableSession | None = None
+        try:
+            session_wrapper = RetryableSession(
+                session_factory=lambda: self._get_driver().session(
+                    database=database, default_access_mode=default_access_mode
+                ),
+                max_retries=SERVICE_UNAVAILABLE_MAX_RETRIES,
+            )
+            yield session_wrapper
+
+        except neo4j.exceptions.Neo4jError as exc:
+            if (
+                default_access_mode == neo4j.READ_ACCESS
+                and exc.code
+                and exc.code in READ_EXCEPTION_CODES
+            ):
+                raise WriteQueryNotAllowedException(
+                    message="Read query not allowed", code=READ_EXCEPTION_CODES[0]
+                )
+
+            message = exc.message if exc.message is not None else str(exc)
+            if exc.code and exc.code.startswith(CLIENT_STATEMENT_EXCEPTION_PREFIX):
+                raise ClientStatementException(message=message, code=exc.code)
+            raise GraphDatabaseQueryException(message=message, code=exc.code)
+
+        finally:
+            if session_wrapper is not None:
+                session_wrapper.close()
+
+    # Operations
+
+    def execute_read_query(
+        self,
+        database: str,
+        cypher: str,
+        parameters: dict[str, Any] | None = None,
+    ) -> neo4j.graph.Graph:
+        with self.get_session(
+            database, default_access_mode=neo4j.READ_ACCESS
+        ) as session:
+
+            def _run(tx: neo4j.ManagedTransaction) -> neo4j.graph.Graph:
+                result = tx.run(
+                    cypher, parameters or {}, timeout=READ_QUERY_TIMEOUT_SECONDS
+                )
+                return result.graph()
+
+            return session.execute_read(_run)
+
+    def create_database(self, database: str) -> None:
+        with self.get_session() as session:
+            session.run(
+                "CREATE DATABASE $database IF NOT EXISTS", {"database": database}
+            )
+
+    def drop_database(self, database: str) -> None:
+        with self.get_session() as session:
+            session.run(f"DROP DATABASE `{database}` IF EXISTS DESTROY DATA")
+
+    def drop_subgraph(self, database: str, provider_id: str) -> int:
+        """Delete all nodes for a provider from a tenant database, batched.
+
+        Deletes relationships then nodes in batches (not `DETACH DELETE`) so a
+        dense provider's graph cannot exceed Neo4j's transaction memory limit.
+        Silently returns 0 if the database doesn't exist.
+        """
+        from api.attack_paths.database import GraphDatabaseQueryException
+        from tasks.jobs.attack_paths.config import (
+            GRAPH_MUTATION_BATCH_SIZE,
+            PROVIDER_RESOURCE_LABEL,
+            get_provider_label,
+        )
+
+        provider_label = get_provider_label(provider_id)
+        deleted_nodes = deleted_relationships = 0
+        relationship_batches = node_batches = 0
+        drop_t0 = time.perf_counter()
+
+        logger.info(
+            "Dropping provider graph from Neo4j sink database %s "
+            "(provider=%s, provider_label=%s)",
+            database,
+            provider_id,
+            provider_label,
+        )
+
+        try:
+            logger.info(
+                "Opening Neo4j sink session for provider graph drop "
+                "(database=%s, provider=%s)",
+                database,
+                provider_id,
+            )
+            with self.get_session(database) as session:
+                logger.info(
+                    "Opened Neo4j sink session for provider graph drop "
+                    "(database=%s, provider=%s)",
+                    database,
+                    provider_id,
+                )
+                log_target = f"Neo4j sink database {database}"
+                for (
+                    phase,
+                    query_template,
+                ) in RELATIONSHIP_DELETE_QUERY_TEMPLATES.items():
+                    deleted_relationships, phase_batches = delete_batches(
+                        session=session,
+                        logger=logger,
+                        log_target=log_target,
+                        provider_id=provider_id,
+                        query=query_template.format(provider_label=provider_label),
+                        phase=phase,
+                        count_key="deleted_rels_count",
+                        total_key="rels",
+                        deleted_key="deleted_rels",
+                        initial_total=deleted_relationships,
+                        batch_size=GRAPH_MUTATION_BATCH_SIZE,
+                        drop_t0=drop_t0,
+                    )
+                    relationship_batches += phase_batches
+
+                deleted_nodes, node_batches = delete_batches(
+                    session=session,
+                    logger=logger,
+                    log_target=log_target,
+                    provider_id=provider_id,
+                    query=NODE_DELETE_QUERY_TEMPLATE.format(
+                        provider_label=provider_label,
+                        provider_resource_label=PROVIDER_RESOURCE_LABEL,
+                    ),
+                    phase="node",
+                    count_key="deleted_nodes_count",
+                    total_key="nodes",
+                    deleted_key="deleted_nodes",
+                    initial_total=0,
+                    batch_size=GRAPH_MUTATION_BATCH_SIZE,
+                    drop_t0=drop_t0,
+                )
+
+        except GraphDatabaseQueryException as exc:
+            if exc.code == DATABASE_NOT_FOUND_CODE:
+                logger.info(
+                    "Skipped provider graph drop from Neo4j sink database %s "
+                    "(provider=%s, reason=database_not_found, elapsed=%.3fs)",
+                    database,
+                    provider_id,
+                    time.perf_counter() - drop_t0,
+                )
+                return 0
+            raise
+
+        logger.info(
+            "Finished dropping provider graph from Neo4j sink database %s "
+            "(provider=%s, relationship_batches=%s, deleted_rels=%s, "
+            "node_batches=%s, deleted_nodes=%s, elapsed=%.3fs)",
+            database,
+            provider_id,
+            relationship_batches,
+            deleted_relationships,
+            node_batches,
+            deleted_nodes,
+            time.perf_counter() - drop_t0,
+        )
+        return deleted_nodes
+
+    def has_provider_data(self, database: str, provider_id: str) -> bool:
+        from api.attack_paths.database import GraphDatabaseQueryException
+        from tasks.jobs.attack_paths.config import (
+            PROVIDER_RESOURCE_LABEL,
+            get_provider_label,
+        )
+
+        provider_label = get_provider_label(provider_id)
+        query = (
+            f"MATCH (n:{PROVIDER_RESOURCE_LABEL}:`{provider_label}`) RETURN 1 LIMIT 1"
+        )
+        try:
+            with self.get_session(
+                database, default_access_mode=neo4j.READ_ACCESS
+            ) as session:
+                result = session.run(query)
+                return result.single() is not None
+
+        except GraphDatabaseQueryException as exc:
+            if exc.code == DATABASE_NOT_FOUND_CODE:
+                return False
+            raise
+
+    def clear_cache(self, database: str) -> None:
+        from api.attack_paths.database import GraphDatabaseQueryException
+
+        try:
+            with self.get_session(database) as session:
+                session.run("CALL db.clearQueryCaches()")
+        except GraphDatabaseQueryException as exc:
+            logger.warning(
+                f"Failed to clear query cache for database `{database}`: {exc}"
+            )
+
+    # Sync write path
+
+    def ensure_sync_indexes(self, database: str) -> None:
+        """Create the `_provider_element_id` lookup index on `_ProviderResource`.
+
+        Every synced node carries the `_ProviderResource` label, so a single
+        index covers both node-upserts and relationship endpoint MATCHes.
+        Without this index the rel sync degrades to a label scan per row and
+        large provider syncs become unworkable.
+        """
+        from tasks.jobs.attack_paths.config import (
+            PROVIDER_ELEMENT_ID_PROPERTY,
+            PROVIDER_RESOURCE_LABEL,
+        )
+
+        query = (
+            f"CREATE INDEX provider_element_id_idx IF NOT EXISTS "
+            f"FOR (n:`{PROVIDER_RESOURCE_LABEL}`) "
+            f"ON (n.`{PROVIDER_ELEMENT_ID_PROPERTY}`)"
+        )
+        with self.get_session(database) as session:
+            session.execute_write(lambda tx: tx.run(query).consume())
+
+    def write_nodes(
+        self,
+        database: str,
+        labels: str,
+        rows: list[dict[str, Any]],
+    ) -> None:
+        if not rows:
+            return
+        from tasks.jobs.attack_paths.config import (
+            PROVIDER_ELEMENT_ID_PROPERTY,
+            PROVIDER_RESOURCE_LABEL,
+        )
+
+        query = f"""
+            UNWIND $rows AS row
+            MERGE (n:`{PROVIDER_RESOURCE_LABEL}` {{`{PROVIDER_ELEMENT_ID_PROPERTY}`: row.provider_element_id}})
+            SET n:{labels}
+            SET n += row.props
+        """
+        with self.get_session(database) as session:
+            session.execute_write(lambda tx: tx.run(query, {"rows": rows}).consume())
+
+    def write_relationships(
+        self,
+        database: str,
+        rel_type: str,
+        provider_id: str,
+        rows: list[dict[str, Any]],
+    ) -> None:
+        if not rows:
+            return
+        from tasks.jobs.attack_paths.config import (
+            PROVIDER_ELEMENT_ID_PROPERTY,
+            PROVIDER_RESOURCE_LABEL,
+            get_provider_label,
+        )
+
+        provider_label = get_provider_label(provider_id)
+        query = f"""
+            UNWIND $rows AS row
+            MATCH (s:`{PROVIDER_RESOURCE_LABEL}`:`{provider_label}` {{`{PROVIDER_ELEMENT_ID_PROPERTY}`: row.start_element_id}})
+            MATCH (t:`{PROVIDER_RESOURCE_LABEL}`:`{provider_label}` {{`{PROVIDER_ELEMENT_ID_PROPERTY}`: row.end_element_id}})
+            MERGE (s)-[r:`{rel_type}` {{`{PROVIDER_ELEMENT_ID_PROPERTY}`: row.provider_element_id}}]->(t)
+            SET r += row.props
+        """
+        with self.get_session(database) as session:
+            session.execute_write(lambda tx: tx.run(query, {"rows": rows}).consume())
+
+    # For compatibility with test harnesses that patch the concrete driver
+    def get_driver(self) -> neo4j.Driver:
+        return self._get_driver()
+
+
+# Helper for tests / external callers that want a writer session specifically
+def get_read_session(
+    sink: Neo4jSink, database: str
+) -> AbstractContextManager[RetryableSession]:
+    return sink.get_session(database, default_access_mode=neo4j.READ_ACCESS)

api/src/backend/api/attack_paths/sink/neptune.py

@@ -0,0 +1,519 @@
+"""AWS Neptune sink implementation.
+
+Dual Bolt drivers: one against the writer endpoint for workers, one against
+the reader endpoint for the API read path. If `NEPTUNE_READER_ENDPOINT` is
+unset the reader falls back to the writer driver so single-node clusters work.
+
+Neptune is single-database. The `database` argument on the SinkDatabase
+protocol is ignored; tenant / provider isolation is enforced by labels that
+the sync step already writes on every node (see tasks/jobs/attack_paths/sync.py).
+
+SigV4 auth lives at the bottom of this file as `neptune_auth_provider`. The
+neo4j driver invokes the returned callable on each token refresh.
+"""
+
+import atexit
+import datetime
+import json
+import logging
+import threading
+import time
+from collections.abc import Callable, Iterator
+from contextlib import contextmanager
+from typing import Any
+from urllib.parse import urlsplit
+
+import neo4j
+import neo4j.exceptions
+from api.attack_paths.retryable_session import RetryableSession, RetryExhaustedError
+from api.attack_paths.sink.base import SinkDatabase
+from api.attack_paths.sink.drop import (
+    NODE_DELETE_QUERY_TEMPLATE,
+    RELATIONSHIP_DELETE_QUERY_TEMPLATES,
+    delete_batches,
+)
+from botocore.auth import SigV4Auth
+from botocore.awsrequest import AWSRequest
+from botocore.session import Session as BotoSession
+from config.env import env
+from django.conf import settings
+from neo4j.auth_management import AuthManagers, ExpiringAuth
+
+logging.getLogger("neo4j").setLevel(logging.ERROR)
+logging.getLogger("neo4j").propagate = False
+
+logger = logging.getLogger(__name__)
+
+SERVICE_UNAVAILABLE_MAX_RETRIES = env.int(
+    "ATTACK_PATHS_SERVICE_UNAVAILABLE_MAX_RETRIES", default=3
+)
+READ_QUERY_TIMEOUT_SECONDS = env.int(
+    "ATTACK_PATHS_READ_QUERY_TIMEOUT_SECONDS", default=30
+)
+# Neptune serverless cold-start can be >30s; give the driver room
+CONN_ACQUISITION_TIMEOUT = env.int("NEPTUNE_CONN_ACQUISITION_TIMEOUT", default=60)
+# TCP connect timeout, ordered below the acquisition timeout so an unreachable
+# endpoint can't pin a request or the readiness probe longer than this. Kept
+# generous: cold-start delays query execution, not the socket connect.
+CONNECTION_TIMEOUT = env.int("NEPTUNE_CONNECTION_TIMEOUT", default=10)
+# Roll connections hourly so SigV4 rotations and cert refreshes don't strand long-lived pool entries
+MAX_CONNECTION_LIFETIME = env.int("NEPTUNE_MAX_CONNECTION_LIFETIME", default=3600)
+MAX_CONNECTION_POOL_SIZE = env.int("NEPTUNE_MAX_CONNECTION_POOL_SIZE", default=50)
+NEPTUNE_WRITE_RETRY_DELAY_SECONDS = 2
+
+READ_EXCEPTION_CODES = [
+    "Neo.ClientError.Statement.AccessMode",
+    "Neo.ClientError.Procedure.ProcedureNotFound",
+]
+CLIENT_STATEMENT_EXCEPTION_PREFIX = "Neo.ClientError.Statement."
+RETRYABLE_WRITE_ERROR_FRAGMENTS = (
+    "Operation failed due to conflicting concurrent operations",
+    "Operation terminated (deadline exceeded)",
+)
+
+# Refresh 60s before the 5-minute SigV4 window closes
+SIGV4_TOKEN_LIFETIME_MINUTES = 4
+
+
+def _is_retryable_write_error(exc: Exception) -> bool:
+    if not isinstance(exc, neo4j.exceptions.Neo4jError):
+        return False
+    message = exc.message or ""
+    return any(fragment in message for fragment in RETRYABLE_WRITE_ERROR_FRAGMENTS)
+
+
+class NeptuneSink(SinkDatabase):
+    """Neptune-backed sink. Single database; isolation is label-based."""
+
+    sync_batch_size = env.int("ATTACK_PATHS_NEPTUNE_SYNC_BATCH_SIZE", default=500)
+
+    def __init__(self) -> None:
+        self._writer: neo4j.Driver | None = None
+        self._reader: neo4j.Driver | None = None
+        self._lock = threading.Lock()
+        self._atexit_registered = False
+
+    # Config
+
+    def _config(self) -> dict:
+        return settings.DATABASES["neptune"]
+
+    def _bolt_uri(self, endpoint: str, port: str) -> str:
+        return f"bolt+s://{endpoint}:{port}"
+
+    def _https_url(self, endpoint: str, port: str) -> str:
+        return f"https://{endpoint}:{port}"
+
+    def _build_driver(self, endpoint: str) -> neo4j.Driver:
+        cfg = self._config()
+        port = cfg["PORT"]
+        region = cfg["REGION"]
+        if not endpoint or not region:
+            raise RuntimeError(
+                "NEPTUNE_WRITER_ENDPOINT and AWS_REGION must be set when "
+                "ATTACK_PATHS_SINK_DATABASE=neptune"
+            )
+        return neo4j.GraphDatabase.driver(
+            self._bolt_uri(endpoint, port),
+            auth=AuthManagers.bearer(
+                neptune_auth_provider(region, self._https_url(endpoint, port))
+            ),
+            keep_alive=True,
+            max_connection_lifetime=MAX_CONNECTION_LIFETIME,
+            connection_timeout=CONNECTION_TIMEOUT,
+            connection_acquisition_timeout=CONN_ACQUISITION_TIMEOUT,
+            max_connection_pool_size=MAX_CONNECTION_POOL_SIZE,
+            max_transaction_retry_time=0,
+        )
+
+    # Lifecycle
+
+    def init(self) -> None:
+        if self._writer is not None:
+            return
+        with self._lock:
+            if self._writer is None:
+                cfg = self._config()
+                writer_endpoint = cfg["WRITER_ENDPOINT"]
+                reader_endpoint = cfg["READER_ENDPOINT"] or writer_endpoint
+
+                # Eager connectivity checks are best-effort
+                # A Neptune that is down at boot must not crash the process, same degradation model as Postgres
+                # Drivers reconnect lazily on first use
+                # /health/ready surfaces the outage until it recovers
+                self._writer = self._build_driver(writer_endpoint)
+                self._verify_best_effort(self._writer, "writer")
+
+                if reader_endpoint == writer_endpoint:
+                    self._reader = self._writer
+
+                else:
+                    self._reader = self._build_driver(reader_endpoint)
+                    self._verify_best_effort(self._reader, "reader")
+
+                if not self._atexit_registered:
+                    atexit.register(self.close)
+                    self._atexit_registered = True
+
+    def close(self) -> None:
+        with self._lock:
+            # `Driver.close()` is idempotent, so closing the same driver twice
+            # (when reader aliases writer on single-endpoint configs) is safe
+            for driver in (self._reader, self._writer):
+                if driver is None:
+                    continue
+                try:
+                    driver.close()
+                except Exception:  # pragma: no cover - best-effort
+                    pass
+            self._writer = None
+            self._reader = None
+
+    # Sessions
+
+    def _get_writer(self) -> neo4j.Driver:
+        self.init()
+        assert self._writer is not None
+        return self._writer
+
+    def _get_reader(self) -> neo4j.Driver:
+        self.init()
+        assert self._reader is not None
+        return self._reader
+
+    @staticmethod
+    def _verify_best_effort(driver: neo4j.Driver, role: str) -> None:
+        try:
+            driver.verify_connectivity()
+
+        except Exception:
+            logger.warning(
+                "Neptune %s endpoint unreachable at init; continuing with a lazily-reconnecting driver",
+                role,
+                exc_info=True,
+            )
+
+    def verify_connectivity(self) -> None:
+        # The API read path uses the reader driver
+        # On single-endpoint clusters it aliases the writer, so this also covers the writer
+        # A writer-only outage is a workers' concern (no HTTP probe there) and deliberately does not fail API readiness
+        self._get_reader().verify_connectivity()
+
+    @contextmanager
+    def get_session(
+        self,
+        database: str | None = None,  # noqa: ARG002 - ignored on Neptune
+        default_access_mode: str | None = None,
+    ) -> Iterator[RetryableSession]:
+        from api.attack_paths.database import (
+            ClientStatementException,
+            GraphDatabaseQueryException,
+            NeptuneWriteRetryExhaustedException,
+            WriteQueryNotAllowedException,
+        )
+
+        driver = (
+            self._get_reader()
+            if default_access_mode == neo4j.READ_ACCESS
+            else self._get_writer()
+        )
+
+        session_wrapper: RetryableSession | None = None
+        try:
+            is_write_session = default_access_mode != neo4j.READ_ACCESS
+            session_wrapper = RetryableSession(
+                session_factory=lambda: driver.session(
+                    default_access_mode=default_access_mode
+                ),
+                max_retries=SERVICE_UNAVAILABLE_MAX_RETRIES,
+                retry_if=_is_retryable_write_error if is_write_session else None,
+                initial_retry_delay_seconds=(
+                    NEPTUNE_WRITE_RETRY_DELAY_SECONDS if is_write_session else 0
+                ),
+                retry_context="Neptune write" if is_write_session else None,
+            )
+            yield session_wrapper
+
+        except RetryExhaustedError as exc:
+            last_error = exc.last_error
+            raise NeptuneWriteRetryExhaustedException(
+                message=str(exc),
+                code=getattr(last_error, "code", None),
+            ) from last_error
+
+        except neo4j.exceptions.Neo4jError as exc:
+            if (
+                default_access_mode == neo4j.READ_ACCESS
+                and exc.code
+                and exc.code in READ_EXCEPTION_CODES
+            ):
+                raise WriteQueryNotAllowedException(
+                    message="Read query not allowed", code=READ_EXCEPTION_CODES[0]
+                )
+
+            message = exc.message if exc.message is not None else str(exc)
+            if exc.code and exc.code.startswith(CLIENT_STATEMENT_EXCEPTION_PREFIX):
+                raise ClientStatementException(message=message, code=exc.code)
+            raise GraphDatabaseQueryException(message=message, code=exc.code)
+
+        finally:
+            if session_wrapper is not None:
+                session_wrapper.close()
+
+    # Operations
+
+    def execute_read_query(
+        self,
+        database: str,  # noqa: ARG002 - ignored on Neptune
+        cypher: str,
+        parameters: dict[str, Any] | None = None,
+    ) -> neo4j.graph.Graph:
+        with self.get_session(default_access_mode=neo4j.READ_ACCESS) as session:
+
+            def _run(tx: neo4j.ManagedTransaction) -> neo4j.graph.Graph:
+                result = tx.run(
+                    cypher, parameters or {}, timeout=READ_QUERY_TIMEOUT_SECONDS
+                )
+                return result.graph()
+
+            return session.execute_read(_run)
+
+    def create_database(self, database: str) -> None:  # noqa: ARG002
+        # Neptune clusters are single-database; there is nothing to create.
+        return None
+
+    def drop_database(self, database: str) -> None:  # noqa: ARG002
+        # Neptune clusters are single-database; there is nothing to drop.
+        return None
+
+    def drop_subgraph(self, database: str, provider_id: str) -> int:  # noqa: ARG002
+        """Delete a provider's subgraph in two bounded phases.
+
+        Neptune write transactions are capped at ~2 minutes. A naive
+        `DETACH DELETE` on a label-scanned batch grows unbounded with graph
+        density (one node can drag thousands of relationships into the same
+        transaction). Instead:
+
+        1. Delete relationships incident to provider nodes, one fixed-size
+           batch per transaction.
+        2. Delete the now-orphaned nodes, one fixed-size batch per transaction.
+
+        Each transaction does work proportional to `batch_size`, never to the
+        graph's branching factor.
+        """
+        from tasks.jobs.attack_paths.config import (
+            GRAPH_MUTATION_BATCH_SIZE,
+            PROVIDER_RESOURCE_LABEL,
+            get_provider_label,
+        )
+
+        provider_label = get_provider_label(provider_id)
+        deleted_relationships = 0
+        relationship_batches = 0
+        node_batches = 0
+        drop_t0 = time.perf_counter()
+
+        logger.info(
+            "Dropping provider graph from Neptune sink "
+            "(provider=%s, provider_label=%s)",
+            provider_id,
+            provider_label,
+        )
+
+        logger.info(
+            "Opening Neptune writer session for provider graph drop (provider=%s)",
+            provider_id,
+        )
+        with self.get_session() as session:
+            logger.info(
+                "Opened Neptune writer session for provider graph drop (provider=%s)",
+                provider_id,
+            )
+            for phase, query_template in RELATIONSHIP_DELETE_QUERY_TEMPLATES.items():
+                deleted_relationships, phase_batches = delete_batches(
+                    session=session,
+                    logger=logger,
+                    log_target="Neptune sink",
+                    provider_id=provider_id,
+                    query=query_template.format(provider_label=provider_label),
+                    phase=phase,
+                    count_key="deleted_rels_count",
+                    total_key="rels",
+                    deleted_key="deleted_rels",
+                    initial_total=deleted_relationships,
+                    batch_size=GRAPH_MUTATION_BATCH_SIZE,
+                    drop_t0=drop_t0,
+                )
+                relationship_batches += phase_batches
+
+            deleted_nodes, node_batches = delete_batches(
+                session=session,
+                logger=logger,
+                log_target="Neptune sink",
+                provider_id=provider_id,
+                query=NODE_DELETE_QUERY_TEMPLATE.format(
+                    provider_label=provider_label,
+                    provider_resource_label=PROVIDER_RESOURCE_LABEL,
+                ),
+                phase="node",
+                count_key="deleted_nodes_count",
+                total_key="nodes",
+                deleted_key="deleted_nodes",
+                initial_total=0,
+                batch_size=GRAPH_MUTATION_BATCH_SIZE,
+                drop_t0=drop_t0,
+            )
+
+        logger.info(
+            "Finished dropping provider graph from Neptune sink "
+            "(provider=%s, relationship_batches=%s, deleted_rels=%s, "
+            "node_batches=%s, deleted_nodes=%s, elapsed=%.3fs)",
+            provider_id,
+            relationship_batches,
+            deleted_relationships,
+            node_batches,
+            deleted_nodes,
+            time.perf_counter() - drop_t0,
+        )
+        return deleted_nodes
+
+    def has_provider_data(self, database: str, provider_id: str) -> bool:  # noqa: ARG002
+        from tasks.jobs.attack_paths.config import (
+            PROVIDER_RESOURCE_LABEL,
+            get_provider_label,
+        )
+
+        provider_label = get_provider_label(provider_id)
+        query = (
+            f"MATCH (n:{PROVIDER_RESOURCE_LABEL}:`{provider_label}`) RETURN 1 LIMIT 1"
+        )
+        with self.get_session(default_access_mode=neo4j.READ_ACCESS) as session:
+            result = session.run(query)
+            return result.single() is not None
+
+    def clear_cache(self, database: str) -> None:  # noqa: ARG002
+        # Neptune has no user-facing cache-clear procedure; no-op.
+        return None
+
+    # Sync write path
+
+    def ensure_sync_indexes(self, database: str) -> None:  # noqa: ARG002
+        # Neptune routes node and relationship lookups through `~id`, which is the cluster's primary key
+        # No additional index is needed or supported
+        return None
+
+    def write_nodes(
+        self,
+        database: str,  # noqa: ARG002
+        labels: str,
+        rows: list[dict[str, Any]],
+    ) -> None:
+        if not rows:
+            return
+        from tasks.jobs.attack_paths.config import (
+            PROVIDER_ELEMENT_ID_PROPERTY,
+            PROVIDER_RESOURCE_LABEL,
+        )
+
+        # MERGE on `~id` is the documented and engine-optimized idempotent
+        # upsert pattern for Neptune openCypher. The label inside the MERGE
+        # matters: Neptune assigns a default `vertex` label to any node
+        # created without an explicit one, so we pin `_ProviderResource`
+        # (which every synced node carries anyway) at MERGE-time. Additional
+        # labels are added after
+        #
+        # We also write `_provider_element_id` as a regular property so
+        # non-sync code (drop_subgraph, query helpers) keeps a stable contract
+        # that doesn't know about `~id`
+        query = f"""
+            UNWIND $rows AS row
+            MERGE (n:`{PROVIDER_RESOURCE_LABEL}` {{`~id`: row.provider_element_id}})
+            SET n:{labels}
+            SET n += row.props
+            SET n.`{PROVIDER_ELEMENT_ID_PROPERTY}` = row.provider_element_id
+        """
+        with self.get_session() as session:
+            session.execute_write(lambda tx: tx.run(query, {"rows": rows}).consume())
+
+    def write_relationships(
+        self,
+        database: str,  # noqa: ARG002
+        rel_type: str,
+        provider_id: str,  # noqa: ARG002 - encoded in start/end `~id` already
+        rows: list[dict[str, Any]],
+    ) -> None:
+        if not rows:
+            return
+        from tasks.jobs.attack_paths.config import PROVIDER_ELEMENT_ID_PROPERTY
+
+        # `id(n) = $value` is Neptune's parameterized fast path; both endpoint
+        # MATCHes resolve in O(1) via the system `~id`, so per-row work stays
+        # bounded regardless of batch size
+        query = f"""
+            UNWIND $rows AS row
+            MATCH (s) WHERE id(s) = row.start_element_id
+            MATCH (e) WHERE id(e) = row.end_element_id
+            MERGE (s)-[r:`{rel_type}` {{`{PROVIDER_ELEMENT_ID_PROPERTY}`: row.provider_element_id}}]->(e)
+            SET r += row.props
+        """
+        with self.get_session() as session:
+            session.execute_write(lambda tx: tx.run(query, {"rows": rows}).consume())
+
+    # Test helpers
+
+    def get_writer(self) -> neo4j.Driver:
+        return self._get_writer()
+
+    def get_reader(self) -> neo4j.Driver:
+        return self._get_reader()
+
+
+# SigV4 auth provider
+
+
+class _NeptuneAuthToken(neo4j.Auth):
+    """Neo4j Auth backed by a SigV4-signed GET to `/opencypher`."""
+
+    def __init__(self, region: str, url: str) -> None:
+        session = BotoSession()
+        credentials = session.get_credentials()
+        if credentials is None:
+            raise RuntimeError(
+                "No AWS credentials available for Neptune SigV4 signing. "
+                "Ensure the boto3 credential chain can resolve."
+            )
+        credentials = credentials.get_frozen_credentials()
+
+        request = AWSRequest(method="GET", url=url + "/opencypher")
+        # SigV4 canonical Host must carry the real `host:port`
+        # Neptune runs on a non-default port (8182), so `.hostname` would drop it and break signing
+        request.headers.add_header("Host", urlsplit(url).netloc)
+        SigV4Auth(credentials, "neptune-db", region).add_auth(request)
+
+        auth_obj = {
+            header: request.headers[header]
+            for header in (
+                "Authorization",
+                "X-Amz-Date",
+                "X-Amz-Security-Token",
+                "Host",
+            )
+            if header in request.headers
+        }
+        auth_obj["HttpMethod"] = "GET"
+
+        super().__init__("basic", "username", json.dumps(auth_obj))
+
+
+def neptune_auth_provider(region: str, https_url: str) -> Callable[[], ExpiringAuth]:
+    """Return a callable the neo4j driver can invoke to refresh credentials."""
+
+    def _provider() -> ExpiringAuth:
+        token = _NeptuneAuthToken(region, https_url)
+        expires_at = (
+            datetime.datetime.now(datetime.UTC)
+            + datetime.timedelta(minutes=SIGV4_TOKEN_LIFETIME_MINUTES)
+        ).timestamp()
+        return ExpiringAuth(auth=token, expires_at=expires_at)
+
+    return _provider

api/src/backend/api/attack_paths/views_helpers.py

@@ -1,12 +1,11 @@
 import logging
-
-from typing import Any, Iterable
+from collections.abc import Iterable
+from typing import Any
 
 import neo4j
-
-from rest_framework.exceptions import APIException, PermissionDenied, ValidationError
-
-from api.attack_paths import database as graph_database, AttackPathsQueryDefinition
+from api.attack_paths import AttackPathsQueryDefinition
+from api.attack_paths import database as graph_database
+from api.attack_paths import sink as sink_module
 from api.attack_paths.cypher_sanitizer import (
     inject_provider_label,
     validate_custom_query,
@@ -16,7 +15,10 @@ from api.attack_paths.queries.schema import (
     RAW_SCHEMA_URL,
     get_cartography_schema_query,
 )
+from api.models import AttackPathsScan
 from config.custom_logging import BackendLogger
+from config.env import env
+from rest_framework.exceptions import APIException, PermissionDenied, ValidationError
 from tasks.jobs.attack_paths.config import (
     INTERNAL_LABELS,
     INTERNAL_PROPERTIES,
@@ -27,6 +29,10 @@ from tasks.jobs.attack_paths.config import (
 logger = logging.getLogger(BackendLogger.API)
 
 
+def _custom_query_timeout_ms() -> int:
+    return env.int("ATTACK_PATHS_READ_QUERY_TIMEOUT_SECONDS", default=30) * 1000
+
+
 # Predefined query helpers
 
 
@@ -103,13 +109,13 @@ def execute_query(
     definition: AttackPathsQueryDefinition,
     parameters: dict[str, Any],
     provider_id: str,
+    scan: AttackPathsScan,
 ) -> dict[str, Any]:
     try:
-        graph = graph_database.execute_read_query(
-            database=database_name,
-            cypher=definition.cypher,
-            parameters=parameters,
-        )
+        # TODO: drop after Neptune cutover
+        # Route reads by the scan row's recorded sink, not by current settings.
+        backend = sink_module.get_backend_for_scan(scan)
+        graph = backend.execute_read_query(database_name, definition.cypher, parameters)
         return _serialize_graph(graph, provider_id)
 
     except graph_database.WriteQueryNotAllowedException:
@@ -143,22 +149,31 @@ def execute_custom_query(
     database_name: str,
     cypher: str,
     provider_id: str,
+    scan: AttackPathsScan,
 ) -> dict[str, Any]:
     # Defense-in-depth for custom queries:
-    # 1. neo4j.READ_ACCESS — prevents mutations at the driver level
-    # 2. inject_provider_label() — regex-based label injection scopes node patterns
-    # 3. _serialize_graph() — post-query filter drops nodes without the provider label
+    # 1. `neo4j.READ_ACCESS` - prevents mutations at the driver level
+    # 2. `inject_provider_label()` - regex-based label injection scopes node patterns
+    # 3. `_serialize_graph()` - post-query filter drops nodes without the provider label
+    # 4. `USING QUERY:TIMEOUTMILLISECONDS` on Neptune - server-side runaway cutoff
     #
     # Layer 2 is best-effort (regex can't fully parse Cypher);
     # layer 3 is the safety net that guarantees provider isolation.
     validate_custom_query(cypher)
     cypher = inject_provider_label(cypher, provider_id)
 
+    # TODO: drop after Neptune cutover
+    backend = sink_module.get_backend_for_scan(scan)
+
+    # Neptune enforces a cluster-level query timeout; prepending the hint
+    # makes the limit explicit and matches the client-side read timeout.
+    # Applies only when the scan's graph lives in Neptune.
+    if getattr(scan, "sink_backend", None) == "neptune":
+        timeout_ms = _custom_query_timeout_ms()
+        cypher = f"USING QUERY:TIMEOUTMILLISECONDS {timeout_ms}\n{cypher}"
+
     try:
-        graph = graph_database.execute_read_query(
-            database=database_name,
-            cypher=cypher,
-        )
+        graph = backend.execute_read_query(database_name, cypher, None)
         serialized = _serialize_graph(graph, provider_id)
         return _truncate_graph(serialized)
 
@@ -181,10 +196,11 @@ def execute_custom_query(
 
 
 def get_cartography_schema(
-    database_name: str, provider_id: str
+    database_name: str, provider_id: str, scan: AttackPathsScan
 ) -> dict[str, str] | None:
     try:
-        with graph_database.get_session(
+        backend = sink_module.get_backend_for_scan(scan)
+        with backend.get_session(
             database_name, default_access_mode=neo4j.READ_ACCESS
         ) as session:
             result = session.run(get_cartography_schema_query(provider_id))

api/src/backend/api/authentication.py

@@ -1,18 +1,19 @@
-from typing import Optional, Tuple
+from math import isfinite
 from uuid import UUID
 
+from api.db_router import MainRouter
+from api.models import TenantAPIKey, TenantAPIKeyManager
 from cryptography.fernet import InvalidToken
+from django.core.exceptions import ObjectDoesNotExist
 from django.utils import timezone
 from drf_simple_apikey.backends import APIKeyAuthentication as BaseAPIKeyAuth
 from drf_simple_apikey.crypto import get_crypto
+from drf_simple_apikey.settings import package_settings
 from rest_framework.authentication import BaseAuthentication
 from rest_framework.exceptions import AuthenticationFailed
 from rest_framework.request import Request
 from rest_framework_simplejwt.authentication import JWTAuthentication
 
-from api.db_router import MainRouter
-from api.models import TenantAPIKey, TenantAPIKeyManager
-
 
 class TenantAPIKeyAuthentication(BaseAPIKeyAuth):
     model = TenantAPIKey
@@ -23,18 +24,49 @@ class TenantAPIKeyAuthentication(BaseAPIKeyAuth):
     def _authenticate_credentials(self, request, key):
         """
         Override to use admin connection, bypassing RLS during authentication.
-        Delegates to parent after temporarily routing model queries to admin DB.
         """
-        # Temporarily point the model's manager to admin database
-        original_objects = self.model.objects
-        self.model.objects = self.model.objects.using(MainRouter.admin_db)
+        try:
+            payload = self.key_crypto.decrypt(key)
+        except ValueError:
+            raise AuthenticationFailed("Invalid API Key.")
+
+        if not isinstance(payload, dict):
+            raise AuthenticationFailed("Invalid API Key.")
+
+        payload_pk = payload.get("_pk")
+        payload_exp = payload.get("_exp")
+        if (
+            not isinstance(payload_pk, str)
+            or isinstance(payload_exp, bool)
+            or not isinstance(payload_exp, (int, float))
+            or not isfinite(payload_exp)
+        ):
+            raise AuthenticationFailed("Invalid API Key.")
 
         try:
-            # Call parent method which will now use admin database
-            return super()._authenticate_credentials(request, key)
-        finally:
-            # Restore original manager
-            self.model.objects = original_objects
+            api_key_pk = UUID(payload_pk)
+        except ValueError:
+            raise AuthenticationFailed("Invalid API Key.")
+
+        if payload_exp < timezone.now().timestamp():
+            raise AuthenticationFailed("API Key has already expired.")
+
+        try:
+            api_key = self.model.objects.using(MainRouter.admin_db).get(id=api_key_pk)
+        except ObjectDoesNotExist:
+            raise AuthenticationFailed("No entity matching this api key.")
+
+        if api_key.revoked:
+            raise AuthenticationFailed("This API Key has been revoked.")
+
+        client_ip = request.META.get(package_settings.IP_ADDRESS_HEADER)
+        if api_key.blacklisted_ips and client_ip in api_key.blacklisted_ips:
+            raise AuthenticationFailed("Access denied from blacklisted IP.")
+
+        if api_key.whitelisted_ips and client_ip not in api_key.whitelisted_ips:
+            raise AuthenticationFailed("Access restricted to specific IP addresses.")
+
+        return api_key.entity, key
 
     def authenticate(self, request: Request):
         prefixed_key = self.get_key(request)
@@ -81,7 +113,7 @@ class CombinedJWTOrAPIKeyAuthentication(BaseAuthentication):
     jwt_auth = JWTAuthentication()
     api_key_auth = TenantAPIKeyAuthentication()
 
-    def authenticate(self, request: Request) -> Optional[Tuple[object, dict]]:
+    def authenticate(self, request: Request) -> tuple[object, dict] | None:
         auth_header = request.headers.get("Authorization", "")
 
         # Prioritize JWT authentication if both are present

api/src/backend/api/base_views.py

@@ -1,3 +1,9 @@
+from api.authentication import CombinedJWTOrAPIKeyAuthentication
+from api.db_router import MainRouter, reset_read_db_alias, set_read_db_alias
+from api.db_utils import POSTGRES_USER_VAR, rls_transaction
+from api.filters import CustomDjangoFilterBackend
+from api.models import Role, UserRoleRelationship
+from api.rbac.permissions import HasPermissions
 from django.conf import settings
 from django.db import transaction
 from rest_framework import permissions
@@ -8,13 +14,6 @@ from rest_framework.response import Response
 from rest_framework_json_api import filters
 from rest_framework_json_api.views import ModelViewSet
 
-from api.authentication import CombinedJWTOrAPIKeyAuthentication
-from api.db_router import MainRouter, reset_read_db_alias, set_read_db_alias
-from api.db_utils import POSTGRES_USER_VAR, rls_transaction
-from api.filters import CustomDjangoFilterBackend
-from api.models import Role, UserRoleRelationship
-from api.rbac.permissions import HasPermissions
-
 
 class BaseViewSet(ModelViewSet):
     authentication_classes = [CombinedJWTOrAPIKeyAuthentication]

api/src/backend/api/compliance.py

@@ -352,7 +352,7 @@ def generate_compliance_overview_template(
                 total_requirements += 1
                 provider_check_list = list(requirement.checks.get(provider_type, []))
                 total_checks = len(provider_check_list)
-                checks_dict = {check: None for check in provider_check_list}
+                checks_dict = dict.fromkeys(provider_check_list)
 
                 req_status_val = "MANUAL" if total_checks == 0 else "PASS"
 

api/src/backend/api/db_utils.py

@@ -2,9 +2,15 @@ import re
 import secrets
 import time
 import uuid
-from contextlib import contextmanager
-from datetime import datetime, timedelta, timezone
+from contextlib import ExitStack, contextmanager, nullcontext
+from datetime import UTC, datetime, timedelta
 
+from api.db_router import (
+    READ_REPLICA_ALIAS,
+    get_read_db_alias,
+    reset_read_db_alias,
+    set_read_db_alias,
+)
 from celery.utils.log import get_task_logger
 from config.env import env
 from django.conf import settings
@@ -22,13 +28,6 @@ from psycopg2 import sql as psycopg2_sql
 from psycopg2.extensions import AsIs, new_type, register_adapter, register_type
 from rest_framework_json_api.serializers import ValidationError
 
-from api.db_router import (
-    READ_REPLICA_ALIAS,
-    get_read_db_alias,
-    reset_read_db_alias,
-    set_read_db_alias,
-)
-
 logger = get_task_logger(__name__)
 
 DB_USER = settings.DATABASES["default"]["USER"] if not settings.TESTING else "test"
@@ -49,6 +48,140 @@ REPLICA_MAX_ATTEMPTS = env.int("POSTGRES_REPLICA_MAX_ATTEMPTS", default=3)
 REPLICA_RETRY_BASE_DELAY = env.float("POSTGRES_REPLICA_RETRY_BASE_DELAY", default=0.5)
 
 SET_CONFIG_QUERY = "SELECT set_config(%s, %s::text, TRUE);"
+SET_TRANSACTION_READ_ONLY_QUERY = "SET TRANSACTION READ ONLY;"
+
+REPLICA_CONNECTION_SQLSTATE_PREFIXES = ("08",)
+REPLICA_CONNECTION_SQLSTATES = {"57P01", "57P02", "57P03"}
+REPLICA_NON_FAILOVER_SQLSTATES = {"57014", "40001", "40P01"}
+REPLICA_CONNECTION_ERROR_MESSAGES = (
+    "ssl syscall",
+    "eof detected",
+    "server closed the connection",
+    "connection already closed",
+    "connection not open",
+    "could not connect to server",
+    "connection refused",
+    "connection reset",
+    "connection timed out",
+    "lost synchronization",
+    "terminating connection",
+    "database system is starting up",
+    "database system is shutting down",
+    "database system is in recovery mode",
+)
+REPLICA_NON_FAILOVER_ERROR_MESSAGES = (
+    "canceling statement due to user request",
+    "deadlock detected",
+    "could not serialize access",
+)
+
+
+def _iter_exception_chain(error: BaseException):
+    seen = set()
+    pending = [error]
+    while pending:
+        current = pending.pop(0)
+        if current is None or id(current) in seen:
+            continue
+        seen.add(id(current))
+        yield current
+
+        cause = getattr(current, "__cause__", None)
+        context = getattr(current, "__context__", None)
+        if cause is not None:
+            pending.append(cause)
+        if context is not None:
+            pending.append(context)
+        for arg in getattr(current, "args", ()):
+            if isinstance(arg, BaseException):
+                pending.append(arg)
+
+
+def _get_exception_sqlstate(error: BaseException) -> str | None:
+    for attr in ("pgcode", "sqlstate"):
+        sqlstate = getattr(error, attr, None)
+        if sqlstate:
+            return sqlstate
+
+    diag = getattr(error, "diag", None)
+    if diag is not None:
+        sqlstate = getattr(diag, "sqlstate", None)
+        if sqlstate:
+            return sqlstate
+    return None
+
+
+def _is_replica_connection_failure(error: BaseException) -> bool:
+    """
+    Return True only for replica failures where retrying on primary is safe.
+
+    Query cancellations, serialization failures, and deadlocks should surface to
+    callers because replaying them can hide real query or concurrency problems.
+    """
+    messages = []
+    sqlstates = set()
+
+    for chained_error in _iter_exception_chain(error):
+        sqlstate = _get_exception_sqlstate(chained_error)
+        if sqlstate:
+            sqlstates.add(sqlstate)
+        messages.append(str(chained_error).lower())
+
+    if sqlstates & REPLICA_NON_FAILOVER_SQLSTATES:
+        return False
+    if any(
+        sqlstate.startswith(REPLICA_CONNECTION_SQLSTATE_PREFIXES)
+        or sqlstate in REPLICA_CONNECTION_SQLSTATES
+        for sqlstate in sqlstates
+    ):
+        return True
+
+    message = " ".join(messages)
+    if any(marker in message for marker in REPLICA_NON_FAILOVER_ERROR_MESSAGES):
+        return False
+
+    return any(marker in message for marker in REPLICA_CONNECTION_ERROR_MESSAGES)
+
+
+def _strip_leading_sql_comments(sql: str) -> str:
+    if not isinstance(sql, str):
+        return ""
+
+    sql_text = sql.lstrip()
+    while True:
+        if sql_text.startswith("--"):
+            newline_index = sql_text.find("\n")
+            if newline_index == -1:
+                return ""
+            sql_text = sql_text[newline_index + 1 :].lstrip()
+            continue
+
+        if sql_text.startswith("/*"):
+            comment_end_index = sql_text.find("*/", 2)
+            if comment_end_index == -1:
+                return ""
+            sql_text = sql_text[comment_end_index + 2 :].lstrip()
+            continue
+
+        return sql_text
+
+
+def _is_safe_primary_replay(sql: str, many: bool) -> bool:
+    if many:
+        return False
+
+    sql_text = _strip_leading_sql_comments(sql)
+    if not re.match(r"(?is)^SELECT\b", sql_text):
+        return False
+
+    return not any(
+        re.search(pattern, sql_text, re.IGNORECASE | re.DOTALL)
+        for pattern in (
+            r"\bINTO\b",
+            r"\bFOR\s+(?:NO\s+KEY\s+)?UPDATE\b",
+            r"\bFOR\s+(?:KEY\s+)?SHARE\b",
+        )
+    )
 
 
 @contextmanager
@@ -78,14 +211,36 @@ def rls_transaction(
     retry_on_replica: bool = True,
 ):
     """
-    Creates a new database transaction setting the given configuration value for Postgres RLS. It validates the
-    if the value is a valid UUID.
+    Context manager that opens an RLS-scoped database transaction.
+
+    Sets a Postgres configuration variable (``set_config``) so that Row-Level
+    Security policies can filter by tenant. When *using* points to a read
+    replica and *retry_on_replica* is True, replica failures are handled in two
+    places:
+
+    1. **Pre-yield** (connection-setup failures): the function retries
+       up to ``REPLICA_MAX_ATTEMPTS`` times on the replica, then falls
+       back to the primary DB.
+    2. **Post-yield** (mid-query failures): an ``execute_wrapper``
+       intercepts connection-level ``OperationalError`` during
+       ``cursor.execute()`` calls and falls back directly to the primary DB
+       for single ``SELECT`` statements. The primary fallback transaction is
+       read-only, and unsafe statements keep raising the original error.
+       The wrapper swaps the inner cursor so ``fetchall()`` / ``fetchone()``
+       read from the new connection transparently.
+
+    Limitation: server-side cursors (``.iterator()``) fetch rows via
+    ``fetchmany()``, which the wrapper does not intercept.  Call sites
+    that iterate large result sets with ``.iterator()`` on the replica
+    should add their own retry logic.
 
     Args:
-        value (str): Database configuration parameter value.
-        parameter (str): Database configuration parameter name, by default is 'api.tenant_id'.
-        using (str | None): Optional database alias to run the transaction against. Defaults to the
-            active read alias (if any) or Django's default connection.
+        value: Database configuration parameter value (must be a valid UUID).
+        parameter: Database configuration parameter name.
+        using: Optional database alias.  Defaults to the active read
+            alias or Django's default connection.
+        retry_on_replica: Whether replica setup failures can retry and
+            connection-level mid-query failures can fall back to primary.
     """
     requested_alias = using or get_read_db_alias()
     db_alias = requested_alias or DEFAULT_DB_ALIAS
@@ -93,54 +248,121 @@ def rls_transaction(
         db_alias = DEFAULT_DB_ALIAS
 
     alias = db_alias
-    is_replica = READ_REPLICA_ALIAS and alias == READ_REPLICA_ALIAS
-    max_attempts = REPLICA_MAX_ATTEMPTS if is_replica and retry_on_replica else 1
+    is_replica = bool(READ_REPLICA_ALIAS and alias == READ_REPLICA_ALIAS)
+    can_failover = is_replica and retry_on_replica
+    replica_alias = alias  # captured before the loop mutates alias
+    max_attempts = (REPLICA_MAX_ATTEMPTS + 1) if can_failover else 1
 
-    for attempt in range(1, max_attempts + 1):
-        router_token = None
-        yielded_cursor = False
+    # State shared between the generator and the _query_failover closure.
+    # The fallback transaction.atomic() is registered into fallback_stack
+    # via enter_context so its __exit__ runs when the outer with-ExitStack
+    # block exits, with the right exc_info. No manual __enter__/__exit__.
+    _fallback = {"succeeded": False, "token": None, "caller_exited_cleanly": False}
 
-        # On final attempt, fallback to primary
-        if attempt == max_attempts and is_replica:
-            logger.warning(
-                f"RLS transaction failed after {attempt - 1} attempts on replica, "
-                f"falling back to primary DB"
-            )
-            alias = DEFAULT_DB_ALIAS
+    with ExitStack() as fallback_stack:
 
-        conn = connections[alias]
-        try:
-            if alias != DEFAULT_DB_ALIAS:
-                router_token = set_read_db_alias(alias)
+        def _query_failover(execute, sql, params, many, context):
+            """execute_wrapper: replay failed replica queries on the primary DB."""
+            try:
+                return execute(sql, params, many, context)
+            except OperationalError as err:
+                if not _is_replica_connection_failure(err):
+                    raise
+                if not _is_safe_primary_replay(sql, many):
+                    raise
 
-            with transaction.atomic(using=alias):
-                with conn.cursor() as cursor:
-                    try:
-                        # just in case the value is a UUID object
-                        uuid.UUID(str(value))
-                    except ValueError:
-                        raise ValidationError("Must be a valid UUID")
-                    cursor.execute(SET_CONFIG_QUERY, [parameter, value])
-                    yielded_cursor = True
-                    yield cursor
-            return
-        except OperationalError as e:
-            if yielded_cursor:
-                raise
-            # If on primary or max attempts reached, raise
-            if not is_replica or attempt == max_attempts:
-                raise
+                try:
+                    connections[replica_alias].close()
+                except Exception:
+                    pass  # Best-effort; connection may already be dead
 
-            # Retry with exponential backoff
-            delay = REPLICA_RETRY_BASE_DELAY * (2 ** (attempt - 1))
-            logger.info(
-                f"RLS transaction failed on replica (attempt {attempt}/{max_attempts}), "
-                f"retrying in {delay}s. Error: {e}"
-            )
-            time.sleep(delay)
-        finally:
-            if router_token is not None:
-                reset_read_db_alias(router_token)
+                logger.warning(
+                    "Mid-query replica connection failure, falling back to primary DB"
+                )
+                primary = connections[DEFAULT_DB_ALIAS]
+                primary.ensure_connection()
+                fallback_stack.enter_context(transaction.atomic(using=DEFAULT_DB_ALIAS))
+
+                fallback_cursor = primary.cursor()
+                fallback_stack.callback(fallback_cursor.close)
+                fallback_cursor.execute(SET_TRANSACTION_READ_ONLY_QUERY)
+                fallback_cursor.execute(SET_CONFIG_QUERY, [parameter, value])
+                _fallback["token"] = set_read_db_alias(DEFAULT_DB_ALIAS)
+
+                fallback_cursor.execute(sql, params)
+
+                context["cursor"].db = primary
+                context["cursor"].cursor = fallback_cursor.cursor
+                _fallback["succeeded"] = True
+                return None
+
+        for attempt in range(1, max_attempts + 1):
+            router_token = None
+            yielded_cursor = False
+
+            # On final attempt, fall back to primary
+            if attempt == max_attempts and can_failover:
+                if attempt > 1:
+                    logger.warning(
+                        f"RLS transaction failed after {attempt - 1} attempts on replica, "
+                        f"falling back to primary DB"
+                    )
+                alias = DEFAULT_DB_ALIAS
+
+            conn = connections[alias]
+            try:
+                if alias != DEFAULT_DB_ALIAS:
+                    router_token = set_read_db_alias(alias)
+
+                with transaction.atomic(using=alias):
+                    with conn.cursor() as cursor:
+                        try:
+                            uuid.UUID(str(value))
+                        except ValueError:
+                            raise ValidationError("Must be a valid UUID")
+                        cursor.execute(SET_CONFIG_QUERY, [parameter, value])
+
+                        wrapper_cm = (
+                            conn.execute_wrapper(_query_failover)
+                            if can_failover and alias == replica_alias
+                            else nullcontext()
+                        )
+                        with wrapper_cm:
+                            yielded_cursor = True
+                            yield cursor
+                            _fallback["caller_exited_cleanly"] = True
+                return
+            except OperationalError as e:
+                if yielded_cursor:
+                    if _fallback["succeeded"] and _fallback["caller_exited_cleanly"]:
+                        # Caller's queries succeeded on primary via failover.
+                        # This error is transaction.atomic() cleanup on the
+                        # dead replica connection, suppress it.
+                        return
+                    raise
+
+                if not can_failover or attempt == max_attempts:
+                    raise
+
+                try:
+                    connections[alias].close()
+                except Exception:
+                    pass  # Best-effort; connection may already be dead
+
+                # Retry with exponential backoff
+                delay = REPLICA_RETRY_BASE_DELAY * (2 ** (attempt - 1))
+                logger.info(
+                    f"RLS transaction failed on replica (attempt {attempt}/{max_attempts}), "
+                    f"retrying in {delay}s. Error: {e}"
+                )
+                time.sleep(delay)
+            finally:
+                if _fallback["token"] is not None:
+                    reset_read_db_alias(_fallback["token"])
+                    _fallback["token"] = None
+
+                if router_token is not None:
+                    reset_read_db_alias(router_token)
 
 
 class CustomUserManager(BaseUserManager):
@@ -170,7 +392,7 @@ def one_week_from_now():
     """
     Return a datetime object with a date one week from now.
     """
-    return datetime.now(timezone.utc) + timedelta(days=7)
+    return datetime.now(UTC) + timedelta(days=7)
 
 
 def generate_random_token(length: int = 14, symbols: str | None = None) -> str:
@@ -405,10 +627,10 @@ def _should_create_index_on_partition(
             # Unknown month abbreviation, include it to be safe
             return True
 
-        partition_date = datetime(year, month, 1, tzinfo=timezone.utc)
+        partition_date = datetime(year, month, 1, tzinfo=UTC)
 
         # Get current month start
-        now = datetime.now(timezone.utc)
+        now = datetime.now(UTC)
         current_month_start = now.replace(
             day=1, hour=0, minute=0, second=0, microsecond=0
         )

api/src/backend/api/decorators.py

@@ -1,14 +1,14 @@
 import uuid
 from functools import wraps
 
-from django.core.exceptions import ObjectDoesNotExist
-from django.db import DatabaseError, connection, transaction
-from rest_framework_json_api.serializers import ValidationError
-
+from api.attack_paths.database import GraphDatabaseQueryException
 from api.db_router import READ_REPLICA_ALIAS
 from api.db_utils import POSTGRES_TENANT_VAR, SET_CONFIG_QUERY, rls_transaction
 from api.exceptions import ProviderDeletedException
-from api.models import Provider, Scan
+from api.models import Membership, Provider, Scan, Tenant
+from django.core.exceptions import ObjectDoesNotExist
+from django.db import DEFAULT_DB_ALIAS, DatabaseError, connection, transaction
+from rest_framework_json_api.serializers import ValidationError
 
 
 def set_tenant(func=None, *, keep_tenant=False):
@@ -76,9 +76,11 @@ def handle_provider_deletion(func):
     """
     Decorator that raises `ProviderDeletedException` if provider was deleted during execution.
 
-    Catches `ObjectDoesNotExist` and `DatabaseError` (including `IntegrityError`), checks if
-    provider still exists, and raises `ProviderDeletedException` if not. Otherwise,
-    re-raises original exception.
+    Catches `ObjectDoesNotExist`, `DatabaseError` (including `IntegrityError`), and
+    `GraphDatabaseQueryException`, checks if provider still exists, and raises
+    `ProviderDeletedException` if not. Graph database errors also check whether the
+    tenant still exists and has memberships. Otherwise, re-raises the original
+    exception.
 
     Requires `tenant_id` and `provider_id` in kwargs.
 
@@ -93,11 +95,16 @@ def handle_provider_deletion(func):
     def wrapper(*args, **kwargs):
         try:
             return func(*args, **kwargs)
-        except (ObjectDoesNotExist, DatabaseError):
+        except (ObjectDoesNotExist, DatabaseError, GraphDatabaseQueryException) as exc:
             tenant_id = kwargs.get("tenant_id")
             provider_id = kwargs.get("provider_id")
+            database_alias = (
+                DEFAULT_DB_ALIAS
+                if isinstance(exc, GraphDatabaseQueryException)
+                else READ_REPLICA_ALIAS
+            )
 
-            with rls_transaction(tenant_id, using=READ_REPLICA_ALIAS):
+            with rls_transaction(tenant_id, using=database_alias):
                 if provider_id is None:
                     scan_id = kwargs.get("scan_id")
                     if scan_id is None:
@@ -114,6 +121,13 @@ def handle_provider_deletion(func):
                     raise ProviderDeletedException(
                         f"Provider '{provider_id}' was deleted during the scan"
                     ) from None
+                if isinstance(exc, GraphDatabaseQueryException) and (
+                    not Tenant.objects.filter(pk=tenant_id).exists()
+                    or not Membership.objects.filter(tenant_id=tenant_id).exists()
+                ):
+                    raise ProviderDeletedException(
+                        f"Tenant '{tenant_id}' was deleted during the scan"
+                    ) from None
             raise
 
     return wrapper