sota-secrets-management

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s AUDIT workflow ingests outsider-authored free text by running secret scanners (e.g., gitleaks detect --source . / trufflehog filesystem . / git log -p) over a repo’s files and commit history, which can include outsider-authored content (issues/PRs/wiki/discussion text or other contributors’ code) that becomes LLM-readable via the agent’s scan/report context.