Stats
Actions
Tags
From claude-evolve
Starts an autonomous evolutionary code optimization run using Claude Code models as mutation operators, iteratively improving code via selection, crossover, and evaluation.
How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-evolve:evolveThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Launch an evolution run that optimizes code using Claude Code models as mutation operators.
Launch an evolution run that optimizes code using Claude Code models as mutation operators.
/evolve [config.json]
/evolve --init initial.py --eval evaluate.py --generations 50
initial.py: Seed program with # EVOLVE-BLOCK-START / # EVOLVE-BLOCK-END markersevaluate.py: Evaluator script that returns metrics and correctnessThe run is autonomous -- it proceeds without interaction until complete or stopped.
npx claudepluginhub samuelzxu/claude-evolve --plugin claude-evolveSocratic interview to clarify evolution tasks before execution. Asks targeted questions across four dimensions, scores ambiguity, produces initial.py, evaluate.py, config.json when ambiguity drops below 20%.
Evolves any measurable artifact (prompt, skill, code) through autonomous mutation-evaluate-gate loops. Supports GT case suites and scalar metric loops with automatic keep/revert.
Runs propose-evaluate-iterate loop to optimize and evolve AI agent performance using LangSmith evaluations and git worktrees for isolation. Requires .evolver.json setup.