Security audit
disease-investigation
Security checks for vulnerabilities and agentic risk
Overview
This skill is a disclosed PatSnap life-science research workflow that requires a user-configured MCP service and does not include hidden code, destructive actions, or unrelated data access.
Before installing, confirm you trust PatSnap with the disease, patent, drug, and business-development queries you submit, protect the API key used in the MCP URL, and treat outputs as research support rather than medical advice.
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Static analysis
No suspicious patterns detected.