docs-ai-prd
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill primarily consists of Markdown templates and reference documentation for product management, technical design, and context engineering. No malicious instructions, DAN patterns, or safety bypass attempts were found.
- [EXTERNAL_DOWNLOADS]: The skill includes a
validate_sources.pyscript used to audit the status of external documentation links. This script performs network requests viaurllibto verify URL availability. The URLs targeted indata/sources.jsonbelong to trusted organizations (Anthropic, GitHub, NIST, OpenAI, OWASP) and well-known services, posing no risk under standard operational use. - [INDIRECT_PROMPT_INJECTION]: The skill provides guidance on specifying AI features and explicitly includes security checklists (
references/security-review-checklist.md) to mitigate prompt injection and data exfiltration risks in the systems being designed. The skill itself does not ingest untrusted external data in an automated way that would compromise the agent's safety.
Audit Metadata