Let the code fix its damn self!
# Agent Fix Agent Fix is a Next.js application that leverages AI-powered agents to identify issues from a GitHub repository and automatically generate code fixes. By integrating with the OpenAI API and GitHub's REST endpoints, Agent Fix analyzes code files, tokenizes and indexes them, and then uses conversational AI to propose fixes along with automated testing. > [!WARNING] > This project will execute code generated by the model on your system. ## Features - **Repository Analysis:** - Input a GitHub repository URL and fetch its open issues. - Automatically download the repository's code files (Python and JavaScript). - **Code Indexing & Embedding:** - Break code files into manageable chunks. - Generate embeddings for each chunk using OpenAI's embeddings API for improved context retrieval. - **AI-Powered Code Fixing:** - Each issue is assigned to an "agent" that generates a potential code fix. - Uses a streaming response to display real-time progress and test execution logs. - Automatically retries with improved suggestions if initial fixes fail the tests. - **Interactive Interface:** - Simple UI built with React and Tailwind CSS. - Allows selection of "reasoning effort" (low, medium, or high) to guide the AI's code-fixing process. - Displays detailed logs, relevant file excerpts, and final code fix results. ## Getting Started ### Prerequisites - Node.js (v14+ recommended) - A GitHub token with permissions to access repository issues and contents. - An OpenAI API key. ### Installation 1. **Clone the Repository** ```bash git clone https://github.com/your-username/agent-fix.git cd agent-fix ``` 2. **Install Dependencies** ```bash npm install ``` 3. **Environment Variables** Create a `.env.local` file in the project root and add the following variables: ```env OPENAI_API_KEY=your_openai_api_key GITHUB_TOKEN=your_github_token ``` 4. **Run the Development Server** ```bash npm
HAL 分层混合模型工作流 — 强模型(Claude)负责理解/拆解/验收,低成本模型(DeepSeek)负责检索/提取/清洗。Hermes Agent skill。
An LLM agent fine-tuned on DeepSeek for spaced repetition, dynamically integrating knowledge points based on the Ebbinghaus forgetting curve.
基于 STM32F103 构建的端到端 AI 智能手表生态。自研“零重定位”原生机器码动态加载引擎与页面栈式 UI 框架;集成生产级 OTA 回滚保护机制与高带宽(921600 baud)串口协议栈。通过 Node.js 中继实现 DeepSeek AI 语义控制及 ASRPRO 语音全双工交互,是一个集成了分布式计算、现代存储管理与 AI Agent 的嵌入式全栈工程。
A Meta-Agent-Driven Self-Evolving Multi-Agent System for UAV Detection and Tracking
One command to run Hermes AI Agent with a browser UI. Zero prerequisites. 一行命令,AI 就位。
网页应用Agent,接入DeepSeek、Mimo等模型