truefoundry-service-test

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (Layer 2 and Layer 3 in SKILL.md) instructs the agent to fetch and parse responses from the deployed service's public endpoints (e.g., https://{HOST}/, /health, and /openapi.json), which are user-provided/untrusted third‑party content that the agent must read as part of its testing flow and that can affect next actions (pass/fail, follow-up tests), exposing it to potential indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly allows registering remote OpenAPI specs and hosted A2A agents that are fetched at runtime and converted into MCP tools or agent behavior (example URLs: "https://api.weather.example.com/openapi.json" and "https://research-agent.example.com/.well-known/agent.json"), and those fetched specs/agent cards can directly control agent capabilities and prompts, creating a high-risk runtime dependency.