prd-review

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection as it ingests and processes untrusted content from user-provided PRD files.
  • Ingestion points: The skill reads local PRD file content based on user input (SKILL.md).
  • Boundary markers: No specific delimiters or instructions are used to isolate the PRD content from the subagent's evaluation logic.
  • Capability inventory: The skill involves reading file content and relaying it to a subagent for natural language analysis (SKILL.md).
  • Sanitization: No validation or sanitization of the document content is performed prior to analysis.
  • [NO_CODE]: The skill consists entirely of markdown instructions and configuration with no executable scripts included.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 10:22 PM
Security Audit — agent-trust-hub — prd-review