
Refact is a self-hosted AI coding assistant for developers who want an agent to work across their repository and development tools. The original project is archived and no longer maintained; ongoing development has moved to JegernOUTT/refact. The original code uses the BSD-3-Clause license.
Its agent can plan coding tasks, make changes, debug problems, and run tests, with developers able to preview and control the work. Repository search gives it project context beyond the current file. It can also draw on documentation and connected databases when working through a task.
Refact integrates with VS Code and JetBrains IDEs, including PyCharm, WebStorm, GoLand, IntelliJ, and CLion. In-editor chat supports code generation, editing, and explanations. Code completion uses Qwen2.5-Coder with retrieval from the project's codebase to suggest lines, functions, and classes as you type.
Connections to GitHub, PostgreSQL, Docker, and CI/CD pipelines let the agent act on development resources as well as source files. Saved use cases and agent memory help retain workflow context across tasks.
On-premise deployment runs the agent on your own infrastructure and supports keeping code on your servers. Refact also supports external model services, including Claude, GPT-4o, Gemini, Grok, and DeepSeek, through API keys. Those services run outside your infrastructure.
Claim this page with an email at refact.ai. Refact gets the verified badge, and you can upgrade the listing to be featured on localhosted. Proud to be listed? Put our badge on your site.
Want more people to find Refact?Promote it
Something wrong or outdated on this page?
28.2KUpdated 1 day agoApache-2.0
macOS · Windows · Linux · Web · VS Code · JetBrains#Code execution#Git integration#MCP
Qwen Code is an Apache 2.0 licensed AI coding agent for developers who want help working through a codebase, changing code and checking the result. It runs on macOS, Windows and Linux, with a terminal interface and a desktop app. It builds on Google Gemini CLI and has developed into an agent that can use Qwen models alongside other model providers.
69.6KUpdated 2 hours agoApache-2.0
macOS · Windows · VS Code · JetBrains#Code execution#Git integration#Human approval
27.5KUpdated 35 minutes agoMIT
VS Code · JetBrains#Code execution#Git integration#MCP
codegpt.coCode Completion and IDE Extensions
VS Code · JetBrains#Code execution#Human approval#LM Studio integration
36.1KUpdated 2 months agoApache-2.0
VS Code · JetBrains#Code execution#Human approval#MCP
29.9KUpdated 1 day ago
Docker · JetBrains#Code execution#MCP#Persistent memory
Cline is an AI coding agent for developers who want help understanding a codebase and changing it. Its shared agent core is open source under Apache 2.0. You can use it in VS Code, JetBrains IDEs, or a terminal. A native desktop app runs on macOS and Windows. No editor is required for desktop sessions. The JetBrains plugin itself is not open source.
Kilo Code is an AI coding agent for developers who want help writing and understanding code in their editor or terminal. It runs in VS Code, JetBrains IDEs and the CLI, with cloud agents available for remote work. The project is open source under the MIT license and can be self-hosted.
CodeGPT is an AI coding assistant for VS Code, JetBrains and Visual Studio that lets developers choose between models running locally through Ollama or LM Studio and cloud models such as Claude, GPT, Gemini and Grok. It's for people who want help writing and debugging code inside their editor while keeping control over which model handles each task.
Continue is an open-source coding agent available as a CLI, VS Code extension and JetBrains plugin. Continue was acquired by Cursor, and the project is no longer actively maintained. Its repository is read-only, and its software uses the Apache 2.0 license.
Serena is a locally run MCP toolkit that gives AI coding agents access to code structure: functions, classes and the references between them. It's for developers who want their agent to find and change specific parts of a codebase without reading whole files or relying on text matching. It requires an LLM client and works with Claude Code, Codex, OpenCode, Cursor and OpenWebUI.