antithesis-documentation

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to fetch and read documentation from the public site https://antithesis.com/docs/ (or via the snouty CLI), so external third‑party docs could contain instructions that influence the agent's decisions and actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs the agent to fetch documentation at runtime from https://antithesis.com/docs/ (including the index https://antithesis.com/docs/llms.txt), and those fetched markdown pages are injected into the model context to drive responses, so this external content directly controls agent prompts.