N/A
humanjudge/HumanJudge
Human-evaluation infrastructure for AI quality. 25,000+ blind human reviews by 200+ verified reviewers across 58 AI models — query the data via five MCP tools (get_model_scores, compare_models, get_flags, check_content, get_latest).
Scan Scheduled
This agent is queued for security scanning. It will be graded in the next scan batch.
What We Know
- URL https://github.com/humanjudge/grandjury
- Framework mcp
- Sources glama
- First Seen May 16, 2026
- Repository github.com/humanjudge/grandjury
Browse more:
Search all agents
Ecosystem Report