forward

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow ("Then: gather cleanup context" in SKILL.md) runs commands like gh pr list and gh issue list which fetch user-generated GitHub PR/issue content (untrusted third‑party data) that the agent is expected to read and which can materially influence planning and actions.