harness-creator
Warn
Audited by Snyk on Jul 1, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.65). At runtime, the skill’s scripts read free-form text from the target repository’s existing harness files (e.g.,
AGENTS.md/CLAUDE.md,progress.md,session-handoff.md,feature_list.json) vialoadHarnessFiles()→readText(), and then include that content in the in-memory strings used for scoring/report generation (and thus into the agent’s LLM context if the agent later summarizes the report); these files are outsider-authored relative to the skill.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata