Run an autonomous coding agent locally using your own API key from a provider like Anthropic, OpenAI, or Groq.
Ask the agent one-shot terminal commands to refactor or explain code without opening a full session.
Build a custom desktop installer of the agent for macOS, Windows, or Linux using the provided build steps.
| adamyasingh-05/codeably | 0xustaz/streamgate | a-bissell/unleash-lite | |
|---|---|---|---|
| Stars | 1 | 1 | 1 |
| Language | Python | Python | Python |
| Setup difficulty | moderate | hard | hard |
| Complexity | 3/5 | 4/5 | 4/5 |
| Audience | developer | developer | researcher |
Figures from each repo's GitHub metadata at analysis time.
Desktop installer needs no setup, but CLI and self-built versions require Python 3.11+, Node 20+, and your own provider API key.
Codeably is an autonomous coding agent that can be downloaded as a desktop app or run from the terminal. It ships with 60 built in tools and can connect to eight different AI providers, including Anthropic, OpenAI, Groq, Gemini, Mistral, Cohere, Ollama, and OpenRouter, so users bring their own API key rather than paying the project directly. The desktop version is distributed as a simple installer for macOS, Windows, and Linux, and requires no Python installation or terminal use once it is set up. On first launch it asks for a one-time email verification, after which the code is stored locally. For developers who prefer the command line, Codeably also offers a terminal based interactive mode and a one-shot command mode where a single instruction can be passed directly, such as asking the agent to refactor a file or explain what a piece of code does. There is also a browser-based interface that can be launched locally without any build step, opening in a normal web browser. Under the hood, the project is built with a shared Python core that holds the tool implementations, the main agent loop, and the provider integrations, alongside a FastAPI backend server and an Electron-based desktop shell with an HTML, CSS, and JavaScript frontend. Developers who want to build the desktop installer themselves can freeze the Python backend into a binary using PyInstaller and then package it with Electron, and the project also supports automatically building and publishing installers through GitHub Actions when a new version tag is pushed. Configuration is handled either through a settings panel inside the app, where API keys are stored locally in the browser, or through standard environment variables like ANTHROPIC_API_KEY and OPENAI_API_KEY. An optional PostgreSQL database can be connected to persist chat sessions and logs of tool usage across runs.
A bring-your-own-key autonomous coding agent with 60 tools, support for 8 AI providers, and desktop, CLI, and browser interfaces.
Mainly Python. The stack also includes Python, FastAPI, Electron.
The README does not state license terms.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.