law-firm-leads
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs a remote Model Context Protocol (MCP) tool from
https://mcp.manypi.com/mcp. This endpoint belongs to the skill's author (manypicom) and provides the backend for the lead generation logic. - [INDIRECT_PROMPT_INJECTION]: The skill identifies several external data sources, such as firm websites and directories, for the agent to process. This creates a surface for indirect prompt injection where malicious instructions hidden in these external sources could influence the agent. * Ingestion points: Regulatory registers, directory profiles, and law firm websites. * Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore embedded commands in the source data. * Capability inventory: The skill uses a custom MCP tool (
manypi) and general web access to retrieve and process lead information. * Sanitization: There are no specified mechanisms to sanitize or validate the content fetched from third-party sites.
Audit Metadata