write-a-prd

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface because it explores the codebase and processes its content to generate a PRD. Malicious instructions embedded in repository files could potentially manipulate the agent's behavior or the output document.
  • Ingestion points: Local repository files (SKILL.md, step 2).
  • Boundary markers: Absent; the skill does not use delimiters or instructions to ignore embedded commands in the data it reads.
  • Capability inventory: File system read (exploring the repo) and file system write (saving the PRD to docs/prds/).
  • Sanitization: Absent; content from the repository is not validated or sanitized before being incorporated into the prompt context.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 08:56 AM