patsnap-biology-modality

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly connects to PatSnap's public MCP server (https://connect.patsnap.com/... and open.patsnap.com) and ingests/searches public patent, literature, and sequence databases as part of its required workflow, which the agent reads and uses to drive analyses and decisions, so untrusted third‑party content could influence actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires connecting at runtime to the PatSnap MCP server (https://connect.patsnap.com/06e741/logic-mcp?apikey=YOUR_API_KEY), which will supply tools/instructions that directly drive the agent's behavior, so the external URL is a runtime dependency that controls prompts.