cli-anything

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

• Third-party content exposure detected (high risk: 1.00). This skill explicitly accepts and clones public GitHub URLs as part of its "Source Acquisition" phase (e.g., the /cli-anything command and the example /cli-anything https://github.com/blender/blender) and then analyzes that codebase to generate CLI harnesses, so untrusted third-party content is ingested and can materially influence agent actions.

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

• Potentially malicious external URL detected (high risk: 0.80). This skill explicitly clones and analyzes external GitHub repositories at runtime (e.g., https://github.com/blender/blender and https://github.com/HKUDS/CLI-Anything), so fetched repo content is required and directly controls the generated CLI prompts/instructions.