shaping
Warn
Audited by Snyk on Feb 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs capturing verbatim "Source" material (user-pasted requests, quotes, emails, Slack messages, etc.) into the Frame/Shaping doc, meaning the agent ingests and displays arbitrary user/third‑party content provided during framing, which could contain indirect prompt-injection.
Audit Metadata