plan-strict

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

• Third-party content exposure detected (high risk: 0.85). The required workflow reads an upstream GitHub issue body (outsider-authored free text) or other external artifacts at runtime in ## Procedure step 2, and then drafts the plan/index/task files from that text, which is fed into the agent’s LLM context during planning.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly supports a GitHub issue input and states "For a GitHub issue, fetch the issue body and title" so it will fetch remote content at runtime that directly controls the plan generation (example URL: https://github.com///issues/).