inspect-change

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from code diffs and requirements documents, creating an attack surface where malicious instructions could influence the agent's review.
  • Ingestion points: The skill ingests data from git diff outputs and external requirements.md files (SKILL.md, Steps 1 and 2).
  • Boundary markers: Explicit boundary markers or instructions to treat diff content as untrusted are absent from the dispatch instructions (SKILL.md, Step 4).
  • Capability inventory: The skill uses git for read-only repository inspection and dispatches subagents to analyze code (SKILL.md, Step 4).
  • Sanitization: There is no mention of sanitizing or escaping the content of the diffs or requirements before processing (SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 03:13 AM
Security Audit — agent-trust-hub — inspect-change