review

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain malicious instructions designed to influence the agent's behavior.\n
  • Ingestion points: The skill reads code diffs (SKILL.md) and pull request/issue comments (references/greptile-triage.md).\n
  • Boundary markers: No explicit delimiters or warnings are used to separate external data from instructions.\n
  • Capability inventory: The skill reports findings and mentions potential code modification capabilities (SKILL.md Guardrails).\n
  • Sanitization: No content validation or sanitization of ingested data is described.\n- [COMMAND_EXECUTION]: The skill utilizes external CLI tools to interact with repository data.\n
  • Evidence: Uses 'gh repo view' and 'gh pr view' to fetch metadata and comments (references/greptile-triage.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 06:25 AM
Security Audit — agent-trust-hub — review