MCP-native AI evaluation: rubric audits, eval suites, and proof reports for AI/LLM output.
Grade, rewrite, and test your AI instructions · Rubrkit reviews prompts, agent specs, commands, skills, workflows, and rubr_flow procedures against a clear 10-dimension rubric, then turns weak instructions into testable rewrites. · An editorial loop for instructions that need to hold. · One grading system for the instructions AI teams actually reuse. · From vague request to testable instruction. · A native format for instructions an agent can actually follow.
| Type | MCP server |
| Section | MCP servers |
| Pricing | free |
| Platform | Command line |
| Systems | cli, api |
| Hosting | cloud |
| Install | mcp |
| Who for | команды и бизнес |
| Protocols | mcp |
| Site language | en |
| GitHub | liberat0r/react-rubrkit |
| Launched | 2026-06-29 |