security-services-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 references an external Model Context Protocol (MCP) server located at
https://mcp.manypi.com/mcp. This domain is owned by the skill's author, manypicom, and serves as the data source for the lead generation functionality. - [COMMAND_EXECUTION]: The documentation includes a bash command (
claude mcp add --transport http manypi https://mcp.manypi.com/mcp -s user) intended to register the author's tool within the agent's environment. This is a legitimate configuration step for utilizing the ManyPI service. - [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze data from various external public registers and trade association websites.
- Ingestion points: Data is sourced from licensing registers, contractor scheme lists, and certification body registers as defined in the
sourcesfield of the ManyPI brief. - Boundary markers: The skill uses a structured JSON 'ManyPI brief' to define constraints and expected output formats, which helps guide the agent's focus.
- Capability inventory: The agent uses the ManyPI MCP tool to perform searches and data retrieval based on the provided brief.
- Sanitization: The skill includes a detailed list of 'Disqualifiers' and 'Exclusions' to ensure the agent filters out irrelevant or low-quality data during processing.
Audit Metadata