review-pr
Fail
Audited by Snyk on Aug 20, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (high risk: 0.80). The skill contains an explicit deceptive instruction to conceal AI authorship ("Do not sign the summary as AI-generated"), which is outside the stated functional purpose of reviewing and merging PRs and therefore is a prompt-injection-like directive to misrepresent provenance.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). In
scripts/review-loop.sh, the required Phase 2/3 runtime repeatedly reads outsider-authored free text atgh api .../issues/$PR/comments,.../pulls/$PR/comments(inline review comments), and.../pulls/$PR/reviews(review summaries) and forwards the emitted[comment] ... @...: <body>lines into the independent triage agent and baseline review prompts.
Issues (2)
E004
CRITICALPrompt injection detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata