jmb-resolving-merge-conflicts
Fail
Audited by Snyk on Nov 19, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (medium risk: 0.30). The skill is a standard git conflict-resolution workflow with no concealment, pretext, or covert code, but its "never --abort" rule and blanket "stage everything and commit" step could push the user into a merge result they did not choose, which is a plausible medium-risk pattern under the recall-first override.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). Step 2 of SKILL.md requires the agent to read GitHub PRs and original issues/tickets, which outsiders can author or comment on without the agent selecting a specific item, so this is medium exposure.
Issues (2)
E004
CRITICALPrompt injection detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata