featbit

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

• Third-party content exposure detected (high risk: 0.80). The skill's mandatory workflow uses webReader to fetch and read real-time content from public sites (docs.featbit.co and GitHub READMEs), so the agent ingests untrusted, user-editable third‑party content that could carry indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's runtime webReader explicitly fetches and injects external documentation into the agent's context (e.g., https://docs.featbit.co and GitHub URLs such as https://github.com/featbit/featbit-dotnet-sdk), so those external pages directly control the agent's responses.