01Integrated Playwright scraper to save offline copies of public ToolBench assessment pages.
02Local web application for interactive access to ToolBench resources and guidance.
03Provides an MCP server ('toolbench_guide') for agents to access ToolBench-related information.
04Offers clear guidance on ToolBench methodology, rescoring processes, and distinctions from other platforms.
05Supports secure and compliant archiving of your own MCP server's assessment reports.
060 GitHub stars