refine-task
Warn
Audited by Snyk on Jul 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: 0.85). Yes—runtime reads of outsider-authored tracker content via
ITEM_URL/ITEM_CONTEXT(e.g., issue bodies/comments) and optional plain web fetches are treated as untrusted data and are ingested into the subagent’s LLM context when assembling the “compact source snapshot” for review.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata