Package Exports

llm-code-fixer
llm-code-fixer/main.js

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (llm-code-fixer) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

LLM Code Fixer

An AI-powered tool that uses vector stores and LLMs (via OpenAI) to automatically detect and fix bugs in JavaScript codebases. It supports syntax, runtime, and logic errors, with an interactive CLI for providing extra context.

Installation

Install globally for CLI access:

npm install -g llm-code-fixer

Set your OpenAI API key:

export OPENAI_API_KEY=your-api-key-here  # On Windows: set OPENAI_API_KEY=your-api-key-here

Usage

Run the CLI to fix a buggy file:

fix-code <path-to-buggy-file>

You'll be prompted for error logs and instructions (optional but recommended for better fixes).
The tool will iteratively fix issues (up to 5 rounds) and output the result.
Backups of original files are saved in your temp directory (e.g., /tmp/llm_fixes or %TEMP%\llm_fixes).

Example

fix-code buggy.js

Paste errors/instructions when prompted, e.g.:

Errors: ReferenceError: x is not defined
Instructions: Add default values for variables.

Inline Usage

Import and use in your code:

import { tryq, fixAndTestFile } from 'llm-code-fixer';

// Wrap a function to auto-fix errors
await tryq(myBuggyFunction);

Dependencies

Required: OpenAI API key.
Optional: faiss-node for faster vector search (npm install faiss-node).

License

MIT

Development

Install Dependencies Locally (for Testing)

In the llm-code-fixer directory:

npm install

Test the CLI Locally

node fix-cli.js <path-to-some-buggy-file>