health-skill-audit

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.65). The required workflow runs python3 scripts/audit-skill-structure.py $ARGUMENTS, and that analyzer is specified to read other skills’ SKILL.md bodies (producing reports like summary.md/overlap-clusters.md); those files are outsider-authored relative to the invoking user when the repo contains third-party skill content, so their free-text SKILL.md content can be ingested by the LLM via the report/summary context path.