anysite-market-research

Pass

Audited by Gen Agent Trust Hub on Oct 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses an execute() meta-tool pattern to interact with an MCP server (anysite). This involves calling endpoints for Y Combinator, SEC, Reddit, LinkedIn, and Twitter/X research. These operations are within the stated purpose of a market research skill.
  • [EXTERNAL_DOWNLOADS]: The skill includes a webparser tool designed to fetch and parse external websites for market intelligence. While this involves downloading external content, it is a primary function of the skill's market intelligence capability and is scoped to research purposes.
  • [SAFE]: All external resources and API patterns trace back to the vendor's own 'anysite' infrastructure and established research platforms (SEC, YC, Reddit). There are no signs of obfuscation, credential theft, or unauthorized persistence mechanisms. The skill follows best practices by using a cache_key system for data management rather than raw sensitive data handling.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 11, 2026, 04:33 AM