darwinian-evolver

Evolutionary search optimizes prompts and regex patterns by iteratively mutating and evaluating them against a fitness function. This Skill uses an evolution loop to evolve code snippets and SQL queries, selecting the most effective solutions over multiple generations.

Yes, you can define a custom problem for code optimization by specifying your own organism, evaluator, and mutator components. This allows you to target specific optimization goals for prompts, regex patterns, SQL queries, or code snippets.

Yes, you need Python >=3.11, git, and an API key for a language model provider to run LLM-driven evolutionary search. These dependencies are required to execute the evolution loop and manage the optimization iterations.

The best way to optimize a regex pattern is to use an evolutionary search loop that evolves the pattern through multiple iterations. By defining an evaluator that tests matching accuracy, the search automatically refines the regex to find the most effective solution.

LLM-driven evolutionary search for SQL queries requires a defined fitness function to evaluate query performance, meaning you must establish clear evaluation metrics. Additionally, optimization quality depends on the language model provider's API and the mutator logic you configure.