skills.rest
mdnaimul22
mdnaimul22COMMUNITY
Explore all skills by this author →
GitHub

openevolve-evolutionary-coding

Run evolutionary coding workflows with deterministic evaluators returning combined fitness scores.

2|Updated Apr 15, 2026
Search Tags:#code optimization#llm agent#fitness evaluation#map-elites#artifact debugging#evolutionary coding#cascade evaluation
One-click install
npx skills add https://github.com/mdnaimul22/human-skills --skill openevolve-evolutionary-coding
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: openevolve-evolutionary-coding
Source: https://github.com/mdnaimul22/human-skills/tree/main/skills/openevolve-evolutionary-coding
Command: npx skills add https://github.com/mdnaimul22/human-skills --skill openevolve-evolutionary-coding

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

OpenEvolve helps you autonomously optimize code when you can define a measurable fitness function, turning slow trial-and-error into evolutionary search driven by LLM mutations and automated evaluators.

Core Features & Use Cases

  • Fitness-driven evolutionary coding: Use an evaluator that returns EvaluationResult(metrics, artifacts) and includes a required combined_score for selection.
  • Multi-stage evaluation (cascade): Run cheap/quick checks first, then only fully evaluate promising candidates.
  • Diverse search with MAP-Elites: Use feature_dimensions derived from evaluator metrics to maintain a varied population of elite solutions.
  • Artifact side-channels for self-correction: Feed stdout/stderr, failure stages, and suggestions back into the next generation to improve iteration quality.
  • Domain coverage: Suitable for evolutionary coding, code optimization, algorithm discovery, and “self-improving” programs across tasks you can automatically grade.

Quick Start

Ask your AI to initialize an OpenEvolve project and then run evolution with your config.yaml, initial_program.py, and a custom evaluator.py that returns combined_score plus optional artifacts.

Frequently Asked Questions about openevolve-evolutionary-coding

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I set up an evaluator for fitness-driven code optimization?▼

MAP-Elites diversity search maintains a varied population of elite solutions by categorizing candidates into feature dimensions derived from evaluator metrics. This ensures the evolutionary search explores diverse program structures rather than converging prematurely.

How do I set up an evaluator for fitness-driven code optimization?▼

Cascade evaluation runs cheap, quick checks first, then fully evaluates only promising candidates. This multi-stage approach saves computational resources by filtering out failing programs before running expensive fitness assessments.

What is MAP-Elites diversity search in evolutionary coding?▼

MAP-Elites diversity search maintains a varied population of elite solutions by categorizing candidates into feature_dimensions derived from evaluator metrics. This ensures the evolutionary search explores diverse program structures rather than converging prematurely.

How does cascade evaluation work in autonomous code optimization?▼

Cascade evaluation runs cheap, quick checks first, then fully evaluates only promising candidates. This multi-stage approach saves computational resources by filtering out failing programs before running expensive fitness assessments.

Can I feed stdout and stderr back into LLM mutations for artifact debugging?▼

Yes, you can feed stdout, stderr, and failure stages back into LLM mutations using artifact side-channels. These artifacts provide error feedback to the next generation, enabling autonomous self-correction and improving iteration quality.

When should I not use evolutionary coding for algorithm discovery?▼

You should not use evolutionary coding for algorithm discovery if you cannot define a deterministic evaluator returning a measurable combined fitness score. The approach requires automated grading to drive the selection of LLM-generated mutations.

Related Skills

Explore More Software Engineering Skills→
yu13130122297yu13130122297

scout

Identify uncertainties in research framing, datasets, metrics, and baselines.

Community
Intermediate
sourcegraphsourcegraph
32

benchmark-audit

Audit benchmark suites against the Agent Benchmark Criteria framework.

Official
Intermediate
Rust-sohamRust-soham

golem-view-agent-logs

Stream agent stdout, stderr, and log channels during invocation.

Community
Basic
ovachieverovachiever
52

stable-baselines3

Train reinforcement learning agents with Stable Baselines3 algorithms.

Community
Advanced