weekly-review

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 external sources that could contain malicious instructions designed to influence the agent's behavior.
  • Ingestion points: The skill reads repository history, project board URLs (e.g., Jira, GitHub), issue descriptions, PR diffs, and CI/monitoring logs as described in the 'Contract' and 'Audit work' sections of SKILL.md.
  • Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore embedded instructions within the processed data.
  • Capability inventory: The skill is authorized to perform scoped source and test edits, publish pull requests, and execute tests via delegates like test-runner and deslop (mentioned in the 'Repair and deslop' section of SKILL.md).
  • Sanitization: The instructions do not include steps to sanitize, escape, or validate the content of external issues or commit messages before using them to drive code repairs or reporting.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 09:24 AM
Security Audit — agent-trust-hub — weekly-review