truefoundry-ai-gateway

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to fetch and ingest remote, user-provided specs and agent cards (see references/manifest-schema.md: "MCP Server (OpenAPI)" remote spec is fetched at runtime and the "Agent" section noting agent_card_url/hosted-a2a-agent are fetched at runtime), which are untrusted third‑party URLs that are converted into MCP tools or agent behavior and can therefore materially influence tool use and decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches external agent and OpenAPI spec URLs at runtime (e.g., https://research-agent.example.com/.well-known/agent.json and https://api.weather.example.com/openapi.json), and those fetched resources are converted into agent cards/MCP tools that directly control agent capabilities and prompts, creating a high-risk runtime dependency.