youskill-finder

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

• Third-party content exposure detected (high risk: 0.80). The skill's documentation and "How It Works" diagram explicitly state the backend performs "SKILL.md Analysis" and uses a "GitHub Fetcher" to read skill documentation from public GitHub repositories, and the CLI displays the returned trigger prompts and skill content—i.e., it ingests and surfaces untrusted, user-generated content from public third-party sources (GitHub) as part of its workflow.

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

• Potentially malicious external URL detected (high risk: 0.90). The CLI makes runtime requests to the YouSkill API at https://labs.youware.com/youskill/api/v1 (see src/api.ts and recommend) and directly injects the returned "trigger_prompt" and skill metadata into its output/agent prompts, so this external endpoint controls prompts/instructions used at runtime.