web-access

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

• Third-party content exposure detected (low risk: 0.10). The skill instructs the agent to fetch external web pages, search results, and social platform posts/comments via search and extraction tools, which ingest outsider-authored free text at runtime.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to fetch installation instructions from an external URL (https://treg.to/llms.txt) during setup, representing an unverifiable third-party operational dependency.