patsnap-chemical-molecular

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

• Third-party content exposure detected (high risk: 0.90). The skill instructs the agent to query the external PatSnap MCP server (e.g., https://connect.patsnap.com/... in SKILL.md/README) and to retrieve and interpret bioactivity data, literature, and patent search results, which are untrusted third‑party sources that can materially influence the agent's analyses and actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires and connects at runtime to the MCP endpoint https://connect.patsnap.com/713886/logic-mcp?apikey=YOUR_API_KEY, which supplies model-context/tools that can directly influence agent prompts and behavior, and is a required external dependency.