{"repo":"skillberry-ai/cap-evolve","free":true,"listed":false,"github":"https://github.com/skillberry-ai/cap-evolve","clone":"git clone https://github.com/skillberry-ai/cap-evolve.git","description":"Optimize any AI agent’s skills, tools/MCP, and prompts against your own evals.","language":"Python","stars":39,"topics":["agent-optimization","agent-skills","agentic-ai","ai-agents","claude","dspy","evals","gepa","llm","llm-evaluation"],"license":"Apache-2.0","category":"ai-agents","readme_excerpt":"cap-evolve watch capability evolve cap-evolve improves an AI agent's prompts, tools, and skills by learning from failed evaluation traces. You bring the agent and the eval you already have. cap-evolve runs the loop — evaluate → diagnose the failures → propose an edit → keep it only if it beats a held-out split by a significant margin → commit — and reports one honest number. It optimizes what your agent reads , not its weights. ▶︎ Watch the 85-second demo · plays inline on the site Quickstart · Documentation · Examples · Results · Contributing Why cap-evolve - Optimize more than prompts. System prompts, executable tool code , MCP tool surfaces, and whole skill packages — pick one or several and optimize them jointly. - Learn from real agent failures. Every iteration reads full trajectories and per-task causal feedback (which task ids a prior edit broke and fixed ), so edits are large and don't regress the wins. - Keep evaluation honest. Acceptance is a val-only significance gate (Δ k·SE); the test split is sealed and scored exactly once. Both live in the core, not in editable docs. - Inspect every change. Each candidate is a git commit; the dashboard shows costs, timing, diffs, lineage, and a tasks × iterations pass/fail heatmap. Try it in two minutes — no API key required toy calc is a deterministic stand-in agent that only answers correctly when its system prompt contains a [CALC] marker. The mock optimizer adds it, so the score provably rises — no model is called . Expecte","default_branch":null,"files":null,"tree":[],"storefront":"/r/skillberry-ai","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/skillberry-ai/cap-evolve/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}