recap
Fail
Audited by Snyk on Nov 17, 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 plain router, but its demo fallback runs unpinned code fetched from a GitHub repo via
npx -y, which is risky without showing malicious intent.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The router itself reads only the user's input and local repo state, but its handoffs to recap-session (git history) and recap-topic (a topic that may require web research) can ingest outsider-authored text that the agent must actively select or that arrives via external contributors' commits, which is at most low exposure.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (medium risk: 0.30). The demo path falls back to
npx -y github:Aboudjem/recap-studio demo, which fetches and executes code from an individual's GitHub repository (a personal, unofficial source with no strong first-party provenance or integrity pinning), so it is a real inbound runtime dependency at the common/reputable unofficial tier.
Issues (3)
E004
CRITICALPrompt injection detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata