projectmanagement
Warn
Audited by Snyk on Mar 9, 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 miniproject skill's Research stage (miniproject/SKILL.md, "Stage: Research") explicitly instructs the agent to use external search/extraction tools (e.g., use_skill(brave_search) and lynx) to fetch web content and says research "informs Spec, Decision, and project direction", meaning untrusted public web content is fetched, read, and can materially influence agent decisions and actions.
Audit Metadata