skills.rest
Signmanal
SignmanalCOMMUNITY
Explore all skills by this author →
GitHub

darwinian-evolver

Automate iterative optimization of prompts, regexes, SQL queries, and code snippets with an LLM-driven evolutionary search loop.

1|Updated Jun 25, 2026
Search Tags:#optimization#llm#sql#evolution#regex#prompt-engineering#code-optimization
One-click install
npx skills add https://github.com/Signmanal/VIGIL --skill darwinian-evolver-signmanal
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: darwinian-evolver
Source: https://github.com/Signmanal/VIGIL/tree/main/optional-skills/research/darwinian-evolver
Command: npx skills add https://github.com/Signmanal/VIGIL --skill darwinian-evolver-signmanal

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires jinja2, openai, and includes scripts (resource) components.

What problem does it solve?

Manually iterating on prompts, regular expressions, SQL queries, and small code snippets to improve performance is slow and relies on trial and error. This skill automates that optimization process using a large language model-driven evolutionary search loop, eliminating repetitive manual tweaking.

Core Features & Use Cases

  • LLM-Driven Evolutionary Optimization: Evolve prompts, regex patterns, SQL queries, and small code snippets against a custom fitness function to maximize performance.
  • Custom Fitness Support: Works with any measurable scorer, including exact match, regex pass rate, unit test results, LLM-as-a-judge scores, and runtime metrics.
  • Flexible Model Access: Includes a built-in OpenRouter driver to use any supported LLM provider, avoiding hardcoded dependencies on specific API services.
  • Use Case Example: A security analyst can use this skill to evolve a prompt template that extracts structured indicators of compromise from unstructured threat reports, improving extraction accuracy over manual iterations.

Quick Start

Use the darwinian-evolver skill to optimize your existing prompt template for verbatim phrase repetition by supplying a set of test input phrases and their expected outputs as the fitness function.

Frequently Asked Questions about darwinian-evolver

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

FAQPage Schema
How do I automate prompt optimization using an LLM?▼

You can automate prompt optimization using an LLM-driven evolutionary search loop that iteratively refines prompt templates against a custom fitness function to maximize performance.

What is evolutionary search for regex and SQL tuning?▼

Evolutionary search for regex and SQL tuning applies an LLM-driven loop to iteratively generate and score candidate patterns or queries against a fitness function, improving them beyond manual trial and error.

Can I use a custom fitness function to score generated code snippets?▼

Yes, you can use any measurable custom fitness function, including exact match, regex pass rate, unit test results, LLM-as-a-judge scores, and runtime metrics to score and evolve code snippets.

Do I need a specific LLM provider API key to run the evolutionary workflow?▼

You need an LLM provider API key to execute the evolutionary workflow, but the built-in OpenRouter driver provides flexible access to any supported LLM provider, avoiding hardcoded API dependencies.

How do I optimize SQL queries with an evolutionary algorithm?▼

You optimize SQL queries by supplying a measurable fitness function that scores query performance, allowing the LLM-driven evolutionary search loop to automatically iterate and refine the queries.

What are the limitations of using LLM-driven evolution for code optimization?▼

This approach requires Python 3.11 or higher, an LLM API key, and the upstream tool installed from GitHub, making it unsuitable for environments lacking these dependencies or without a measurable fitness function.

Related Skills

Explore More Software Engineering Skills→
xukrutdonutxukrutdonut

sql-gen

Translate natural language prompts into executable SQL queries across multiple dialects.

Community
Advanced
519lab519lab
2

darwinian-evolver

Evolve prompts, regex patterns, SQL queries, and code snippets via an LLM-driven loop.

Official
Advanced
heysuhasheysuhas
2

darwinian-evolver

Evolve prompt templates, regex patterns, and code snippets via LLM-driven search.

Community
Advanced
zangjeicyzangjeicy

darwinian-evolver

Evolve prompts, regex patterns, SQL queries, and code snippets via LLM-driven mutation and fitness scoring.

Community
Advanced