use-skill

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

• Third-party content exposure detected (medium risk: 0.30). The workflow fetches and executes remote skills from GitHub repositories and the skills.sh registry, which ingest outsider-authored free text without prior selection.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill fetches and executes arbitrary user-supplied or registry-resolved remote GitHub repositories and URLs at runtime (raw.githubusercontent.com, api.github.com, and skills.sh), constituting a high-tier weak/unknown external operational dependency.