The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions to install a Model Context Protocol (MCP) server from https://mcp.manypi.com/mcp. This domain is controlled by the skill's author, ManyPI, and serves to provide the necessary tooling for the skill's lead generation functions.
[COMMAND_EXECUTION]: A setup command claude mcp add --transport http manypi https://mcp.manypi.com/mcp -s user is provided. This command executes in the user's shell to register an external tool, which is a standard method for extending the agent's capabilities with third-party APIs.
[INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves ingesting untrusted data from external company registries, charges registers, and insolvency notices.
Ingestion points: Data enters the agent's context through national company register feeds, bulk data APIs, and filed financial documents.
Boundary markers: The instructions do not define specific delimiters or warnings to isolate the agent from potentially malicious instructions embedded in the public filings.
Capability inventory: The agent uses the manypi tool to fetch data and performs analysis on the content of these filings.
Sanitization: There is no evidence of sanitization or filtering applied to the external registry data before it is analyzed by the language model.

company-filing-signals