darwinian-evolver

You can optimize prompts by running an LLM-driven evolutionary loop that iteratively mutates and evaluates prompt variations against a custom fitness function. This Skill automates the prompt optimization process, generating and testing candidate prompts to maximize your defined evaluation metrics.

Evolutionary search for code optimization works by treating code snippets as organisms, applying LLM-driven mutators to generate variations, and scoring them with evaluators against a fitness function. The evolutionary loop iteratively selects and mutates the fittest code snippets to improve performance or accuracy.

Yes, you need Python version 3.11 or higher, git, and a language model provider API key to run the evolutionary search. These prerequisites are strictly required to execute the LLM-driven evolutionary loop and manage the optimization environment.

Yes, you can define a custom fitness function to evaluate and optimize SQL queries by specifying your own organisms, evaluators, and mutators. The evolutionary search loop will use your custom evaluator to score and iteratively refine the SQL queries for better performance.

The best way to automate prompt engineering is using an LLM-driven evolutionary loop that systematically mutates and evaluates prompts against a target fitness function. This approach automates the refinement process, allowing you to optimize inputs without manual trial and error.