insurance-broker-leads
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the user to execute a bash command (
claude mcp add) to add a new Model Context Protocol (MCP) server to the agent's environment. - [EXTERNAL_DOWNLOADS]: The skill relies on a remote MCP server hosted at
https://mcp.manypi.com/mcp. This is a vendor-owned resource associated with the skill's author, manypicom. - [INDIRECT_PROMPT_INJECTION]: The skill functions by retrieving and analyzing data from various external sources, which introduces a surface for indirect prompt injection attacks.
- Ingestion points: Regulatory registers, broker association directories, company websites, and company filings.
- Boundary markers: The skill does not define clear delimiters or use specific instructions to prevent the agent from following commands embedded in the external data.
- Capability inventory: The skill uses the ManyPI MCP server to interact with and process data from the identified sources.
- Sanitization: No explicit sanitization or validation of the retrieved content is mentioned before it is presented to the agent.
Audit Metadata