refactor-understand
Warn
Audited by Snyk on Jul 6, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.70). The required runtime workflow reads the target repository’s files (e.g.,
package.json, lockfiles, and especially.github/workflows/*and other repo text) into the LLM context via the harnessnode .../diagnose.mjs signals --target <dir>, so any of that repo’s free-form text authored by an outsider can be ingested (e.g., CI config/README-like content extracted as prose).
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata