context-injection
Warn
Audited by Snyk on Apr 22, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.70). The SKILL.md explicitly instructs the agent to ingest and inject "retrieved chunks" (e.g., code showing retrieve(userMessage) and an example context with a public URL like https://stripe.com/docs/webhooks) into prompts so the model will read and use external public/third‑party content as part of its workflow, which could enable indirect prompt injection.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata