AuditArchitect OS eligible

RAG Audit

Audits RAG systems for compliance with retrieval governance. Checks retrieval order, prompt hardening, injection defense, context firewall, metadata provenance, hybrid retrieval, and code generation grounding.

How to activate

Say any of these phrases in your IDE to trigger this skill:

“RAG audit”“audit retrieval”“check RAG”“retrieval governance”“prompt injection check”

Run via CLI

enterprise-skills run rag-audit

Aliases:

audit-retrievalcheck-ragretrieval-governance

Relationships

Depends on:

Governance Enforcement

Enforces Pre-Code Gate, model roles, preflight resolution, mid-session handoff, execution logging, and feedback capture. Reads .project-ai/ authority pack when available.

Coordinates with:

Full Stack Audit

Runs ALL enterprise audit skills in sequence — security, code review, API contracts, dependency audit, pre-deploy, testing verification.

Security Audit

Scans for auth gaps, hardcoded secrets, missing RLS, XSS, CSRF, unprotected routes, exposed env vars.

Anti-Hallucination Agent

Deep anti-hallucination verification aligned with Anthropic and OpenAI best practices. Cross-reference validation, code grounding checks, capability claim verification, RAG grounding, and refactor integrity analysis. Three-layer enforcement: baseline governance, code generation guards, and on-demand deep audit.