mcp-builder
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses WebFetch to load official documentation and SDK reference materials from trusted domains, specifically modelcontextprotocol.io and GitHub repositories associated with the Model Context Protocol organization.
- [COMMAND_EXECUTION]: The evaluation framework (
scripts/evaluation.py) leverages standard input/output (stdio) to launch and interact with MCP server processes locally for testing purposes, which is the intended functionality of the tool. - [SAFE]: No malicious patterns, obfuscation, or sensitive data exposure were detected. The skill actively promotes secure development practices, such as input validation via Pydantic/Zod schemas and the use of environment variables for credential management.
Audit Metadata