consensus
Warn
Audited by Snyk on Jul 29, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). In SKILL.md Step 2-3, the outsider-authored
<prompt>(user free text) is embedded verbatim into the per-pass prompt sent to each LLM sub-agent, so arbitrary user text is directly ingested without preselecting any specific item.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata