batch-refine
Warn
Audited by Snyk on May 11, 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 explicitly ingests user-generated GitHub content (it uses GitHub MCP / gh to list merged PRs with the knowledge:pending label, read PR details/authors, and re-read the Report PR progress table) and uses those PR contents/outputs (e.g., sufficiency verdicts, evidence bundles, author resolution) to decide pipeline actions like candidate acceptance, label flips, commits, and retriggers — exposing the agent to untrusted third‑party content that can materially influence behavior.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata