feat(api): make Attack Paths sink selectable between Neo4j and Neptune

.gitignore

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

README.md

@@ -83,16 +83,35 @@ prowler dashboard
 
 ## 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.
 

api/CHANGELOG.md

@@ -2,6 +2,14 @@
 
 All notable changes to the **Prowler API** are documented in this file.
 
+## [1.32.0] (Prowler UNRELEASED)
+
+### 🔄 Changed
+
+- Attack Paths: AWS Neptune is now supported as a persistent sink database, selectable via `ATTACK_PATHS_SINK_DATABASE=neptune` (default `neo4j`), Cartography's per-scan staging database stays on Neo4j [(#11524)](https://github.com/prowler-cloud/prowler/pull/11524)
+
+---
+
 ## [1.31.0] (Prowler UNRELEASED)
 
 ### 🚀 Added
@@ -74,7 +82,7 @@ All notable changes to the **Prowler API** are documented in this file.
 ### 🚀 Added
 
 - GIN index on `findings(categories, resource_services, resource_regions, resource_types)` to speed up `/api/v1/finding-groups` array filters [(#11001)](https://github.com/prowler-cloud/prowler/pull/11001)
-- `GET /health/live` and `GET /health/ready` Kubernetes-style probe endpoints following the IETF Health Check Response Format (`application/health+json`). Readiness verifies PostgreSQL, Valkey and Neo4j connectivity and returns 503 with per-dependency detail when any is unreachable [(#11200)](https://github.com/prowler-cloud/prowler/pull/11200)
+- `GET /health/live` and `GET /health/ready` Kubernetes-style probe endpoints following the IETF Health Check Response Format (`application/health+json`). Readiness verifies PostgreSQL, Valkey and Neo4j connectivity and returns 503 with per-dependency detail when any is unreachable; both endpoints centralize the API version on `config/version.py` (read from `pyproject.toml`) and are wired into the Helm charts and the Docker Compose healthcheck [(#11200)](https://github.com/prowler-cloud/prowler/pull/11200)
 
 ### 🔄 Changed
 

api/pyproject.toml

@@ -56,7 +56,7 @@ dependencies = [
   "matplotlib (==3.10.8)",
   "reportlab (==4.4.10)",
   "neo4j (==6.1.0)",
-  "cartography (==0.135.0)",
+  "cartography (==0.136.0)",
   "gevent (==25.9.1)",
   "werkzeug (==3.1.7)",
   "sqlparse (==0.5.5)",
@@ -174,7 +174,7 @@ constraint-dependencies = [
   "blinker==1.9.0",
   "boto3==1.40.61",
   "botocore==1.40.61",
-  "cartography==0.135.0",
+  "cartography==0.136.0",
   "celery==5.6.2",
   "certifi==2026.1.4",
   "cffi==2.0.0",
@@ -425,7 +425,7 @@ 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",
@@ -436,7 +436,7 @@ 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
+# prowler@master needs okta==3.4.2; cartography 0.136.0 declares okta<1.0.0 for an
 # integration prowler does not import.
 #
 # prowler@master hard-pins microsoft-kiota-abstractions==1.9.2 in [project.dependencies].

api/src/backend/api/apps.py

@@ -42,9 +42,6 @@ def ready(self):
         ):
             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/cypher_sanitizer.py

@@ -4,10 +4,10 @@
 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:
@@ -27,24 +27,24 @@
 
 
 # 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",
     re.IGNORECASE,
 )
 
 # 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 +57,9 @@
     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)\(" r"(\s*[a-zA-Z_]\w*)" r"(\s*(?:\{[^}]*\})?)" r"\s*\)"
 )
@@ -136,9 +136,7 @@ def _save(match):
     return work
 
 
-# ---------------------------------------------------------------------------
 # Validation
-# ---------------------------------------------------------------------------
 
 # Patterns that indicate SSRF or dangerous procedure calls
 # Defense-in-depth layer - the primary control is `neo4j.READ_ACCESS`