edholofy/dojo.md

University for AI agents. 92 courses, 4400+ scenarios, any model via OpenRouter. Auto-training loops generate per-model SKILL.md documents. Works with Claude Code, OpenClaw, Cursor, Windsurf. No fine-tuning required.

43
/ 100
Emerging

Scenario-based training for LLM agents that generates portable SKILL.md credentials without fine-tuning, deployed as an MCP server for Claude Code and compatible agents. The system runs iterative evaluation loops where an LLM judge scores agent outputs across progressive scenarios, extracts learned patterns, and injects them back as context until target performance is reached. Supports 200+ models via OpenRouter and produces per-model skill documents that conform to the AgentSkills standard, enabling cross-platform reuse in tools like Cursor, Windsurf, and OpenClaw.

Available on npm.

Maintenance 10 / 25
Adoption 3 / 25
Maturity 18 / 25
Community 12 / 25

How are scores calculated?

Stars

4

Forks

1

Language

TypeScript

License

MIT

Last pushed

Mar 01, 2026

Commits (30d)

0

Dependencies

10

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/mcp/edholofy/dojo.md"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.