swe-reviewer

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes data from external sources like pull request descriptions and issue comments.
  • Ingestion points: Processes untrusted content via vcs-issue-management and vcs-change-request skills as described in SKILL.md.
  • Boundary markers: The workflow lacks explicit boundary markers or instructions to ignore embedded commands in external data.
  • Capability inventory: The skill can create pull requests and update issue checklists.
  • Sanitization: No sanitization or validation of external input is specified.
  • Mitigation: The skill incorporates a human-in-the-loop validation step (MITM Validation) before finalized delivery.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 02:27 AM