Large Language Maid

llmaid is a command-line tool designed to automate the process of AI supported file changes using large language models. It reads source code files, sends them to a Ollama or an OpenAI-compatible API, and writes back the models answers. The tool is highly configurable and supports every kind of text-based input file.

Note

1. Paid services such as ChatGPT can cause high API costs if they are used with many files. Double check your jobs.
2. You may get lower quality when using local models with Ollama, but it's completely free and your files will never leave your computer.

What can it do?

llmaid will run through every file in a path you specify and rewrite, analyse or summarize it. Pretty much everything you can come up with, as long as you can write a good system prompt.

Note

The quality mainly depends on the model you want to use. ChatGPT works great, but I have also had good results with medium sized local models like mistral-small:22b and qwen2.5:32b.

This repository provides a few system prompts, for example:

Documenting code

With this prompt, llmaid will scan and rewrite each code file and generate missing summaries, fix typos, remove wrong comments and much more:

Finding unprofessional slang

This prompt will output one json code block for each file. There it lists findings such as insults, cringe comments, and much more including a severity rating and a description what it thinks about the things it found:

Prerequisites

• .NET 8.0 SDK
• An Ollama instance or an OpenAI-compatible API including api key

Installation

git clone https://github.com/awaescher/llmaid
cd llmaid

# edit ./llmaid/appsettings.json to your needs
# edit ./systemprompt.txt to your needs

dotnet run --project llmaid

Configuration

Change the appsettings.json file in the root directory to your needs:

{
  "Provider": "ollama",                        // ollama or openai (works with any compatible api)
  "Uri": "https://localhost:11434",            // Ollama or OpenAI (compatible) endpoints like http://localhost:11434 or https://api.openai.com
  "ApiKey": ""                                 // not required for Ollama
  "Model": "deepseek-coder-v2:16b",            // the model to use
  "Mode": "replacefile",                       // "replacefile" tries to rewrite files while "find" just writes console outputs - both depending on the system prompt
  "PromptFile": "./systemprompt.txt",          // the system prompt to prime the model
  "SourcePath": "./testcode",                  // the path to look for files to change
  "FilePatterns": [ "*.cs", "*.js" ],          // the file types to change
  "Temperature": 0.7,                          // the models temperature (0 precise to 1 creative)
  "WriteCodeToConsole": true,                  // whether or not the models response should be shown in the console
  "ReplaceFiles": true                         // whether or not the files should be replaced with the model's response
}

FAQ

I get an 404 (Not Found)

It is very likely that Ollama returns this 404 because it doesn't know the model that's specified in the appsettings.json. Make sure to specify a model you have downloaded.

---

❤ Made with Spectre.Console and OllamaSharp.