security-advisory-review

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs the agent to "trigger when the user shares a security advisory" and to read/evaluate advisories, PoCs, and GHSA/CVE reports, so the skill ingests untrusted, user/third‑party advisory content (from pasted reports or public disclosures) that can materially influence its assessment and next actions.