decision-debate
Warn
Audited by Snyk on Aug 5, 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, the Stage 1 “independent, blind research” for each persona explicitly uses WebSearch (plus a knowledge MCP and the MEL Wiki), meaning outsider-authored free text can be ingested from search results/pages at runtime unless additional constraints prevent it.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata