specify-behavior

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external, potentially untrusted project data (such as frame-change outcomes and various system documents) and interpolates them into the generated requirements.md file.
  • Ingestion points: SKILL.md describes loading data from docs/product/metrics.md, docs/security/threat-model.md, docs/ops/reliability.md, and other system docs to ground non-functional requirements (NFRs).
  • Boundary markers: The skill does not define specific delimiters or "ignore" instructions when including content from these documents into the requirements file.
  • Capability inventory: The skill performs file reading (grep/read), file writing (creating requirements.md in docs/specs/), and directory management (moving folders in .skills/).
  • Sanitization: While the skill includes a "Self-review" step with ambiguity and testability scans, it lacks automated sanitization or escaping of the interpolated external data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 03:13 AM
Security Audit — agent-trust-hub — specify-behavior