patent-trademark-filing-leads
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to add an MCP (Model Context Protocol) tool from a remote URL (
https://mcp.manypi.com/mcp). This URL belongs to the skill author and is used to provide the primary functionality of the skill. - [COMMAND_EXECUTION]: The documentation provides a bash command (
claude mcp add ...) to register a remote tool server with the agent environment. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from untrusted external sources, such as public trademark registers and company websites, which presents an attack surface for indirect prompt injection. Ingestion points: External IP registries and company websites. Boundary markers: None identified. Capability inventory: The skill utilizes an external tool (ManyPI) to process data. Sanitization: No specific sanitization or filtering logic is described in the prompt instructions.
Audit Metadata