pr-shepherd
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external GitHub sources, such as pull request comments, review threads, and CI logs, which constitutes a potential surface for indirect prompt injection attacks. However, the risk is mitigated by the skill's specific operational scope and logic constraints.\n
- Ingestion points: Pull request metadata, review comments, and CI logs are fetched using
gh pr view,gh api, andgh run view.\n - Boundary markers: The skill does not define specific delimiters to isolate external data from its primary instructions, although it provides clear logic for handling different PR states.\n
- Capability inventory: Capabilities include repository interactions via the
ghtoolset, local file modifications for bug fixes, and process monitoring.\n - Sanitization: The instructions do not specify sanitization or escaping of external content before processing, relying instead on agent-level guardrails.
Audit Metadata