mcp-builder
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches documentation and SDK information from official Model Context Protocol resources, including the documentation site (modelcontextprotocol.io) and GitHub repositories (github.com/modelcontextprotocol). These are used to provide the agent with the latest protocol specifications and development guidelines.
- [COMMAND_EXECUTION]: The evaluation script (
scripts/evaluation.py) is designed to launch and interact with local MCP servers. This involves executing local commands (e.g.,python,node, ornpx) to run the server or testing tools. This is a legitimate and necessary function for server development and verification. - [DATA_EXFILTRATION]: The evaluation harness communicates with the Anthropic API and remote MCP servers (via SSE or HTTP) to conduct tests. These network operations are part of the core functionality for evaluating the server's tools and responses.
Audit Metadata