mcp-builder
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a documentation and tooling package for MCP development. It provides reference guides and scripts for building and testing MCP servers.
- [EXTERNAL_DOWNLOADS]: The skill references official MCP repositories and documentation. All remote references target the 'modelcontextprotocol' organization on GitHub or 'modelcontextprotocol.io', which are trusted sources for this technology.
- [COMMAND_EXECUTION]: The skill includes an evaluation script (
scripts/evaluation.py) that executes user-specified commands to launch MCP servers for testing. This is the intended purpose of the evaluation harness and is performed locally under the user's control. - [DATA_EXfiltration]: No data exfiltration patterns were detected. The scripts interact with the Anthropic API and specified MCP servers for the purpose of evaluation.
- [CREDENTIALS_UNSAFE]: The skill does not contain hardcoded credentials. It uses environment variables (e.g.,
ANTHROPIC_API_KEY) for authentication, which is a standard and safe practice.
Audit Metadata