deep-resolve
Warn
Audited by Snyk on Apr 10, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). The skill's Step 4 and Tool Usage explicitly require web searches and GitHub/Reddit/YouTube lookups ("Web search", "gh search", and the recommended
agent-reach/ddg-search) and instruct the agent to read and validate external public sources and incorporate their findings into recommendations, which clearly ingests untrusted, user-generated third‑party content that can influence actions.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata