mcp-builder
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The evaluation script (
scripts/evaluation.py) and connection utility (scripts/connections.py) allow the agent to execute local commands and arguments using thestdiotransport. This is the primary intended functionality of the skill, designed to launch and test locally developed MCP servers during the development lifecycle. - [EXTERNAL_DOWNLOADS]: The skill documentation suggests fetching official protocol specifications and SDK documentation from the Model Context Protocol organization on GitHub. Additionally, the evaluation harness interacts with the Anthropic API to facilitate automated testing. These interactions are directed at well-known and established services.
- [SAFE]: The skill demonstrates adherence to security best practices, such as instructing users to manage sensitive credentials like
ANTHROPIC_API_KEYthrough environment variables. No evidence of malicious obfuscation, prompt injection, or unauthorized data exfiltration was found across the analyzed files.
Audit Metadata