Claude Code skills for building LLM evals
Structured eval workflow for Claude Code when LangSmith and Braintrust already exist.
A multi-model workflow for choosing open-source project ideas that fit your background and career goals.
Multi-model debate workflow for OSS ideas, but it's sophisticated prompt chaining.
Developers looking for open-source project ideas
Cursor · GitHub Copilot · ChatGPT
Structured eval workflow for Claude Code when LangSmith and Braintrust already exist.
Tests workflow + tool + model together, not just model capability like SWE-bench.
MCP-enabled workflow orchestrator, but tied to one person's ecosystem limits adoption.
Audit-ready AI agent that replays verified workflows instead of re-reasoning every time.
Multi-agent critique that argues with your idea before /scaffold hands it to Claude Code.
Enterprise agent IDE with evals and observability, but LangChain, LlamaIndex, and Qdrant already own this.