bootstrap

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly clones and inspects arbitrary public GitHub repositories provided via the required "Repository URL" input and Procedure step 1 in SKILL.md, then reads repo files to detect toolchains and run install/start commands, which exposes the agent to untrusted, user-generated content that can materially influence tool use and actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill requires a repository URL at runtime and explicitly clones and runs project setup/verification commands from it (e.g., https://github.com/owner/repo), so fetched repo content can execute remote code and is a required runtime dependency.