An open-source coding agent that reads your project, edits files and runs tests from your terminal, desktop, browser or editor.
clacky-ai/openclacky
An open-source AI agent in Ruby that runs coding and general tasks with a minimal tool set, using any OpenAI-compatible model.
Open-source alternative to:Claude Code
- Stars
- 1,206
- Forks
- 113
- Open issues
- 28
- Last commit
- 15 hours ago
What is it?
OpenClacky is an open-source AI agent that can be used from a terminal or a local Web UI for coding, writing, and research sessions. It works with any OpenAI-compatible API, with built-in setup for Claude, GPT, DeepSeek, Kimi, MiniMax, and OpenRouter, and users supply their own API keys.
The project focuses on token efficiency. It exposes 16 core tools and offloads other capabilities to a Skill system, where skills are invoked with a slash command, can be drafted from natural language as SKILL.md files, and can be updated after each run. It also compresses long context during idle time to keep the prompt cache warm, and supports separate permission modes for manual or automatic tool approval.
Installation is available as a desktop installer for macOS and Windows, as a RubyGem (Ruby 3.1 or newer), through shell scripts, or as a Docker image. The Web UI supports multiple parallel sessions and can be protected with an access key.
Key features
- 01Runs in a terminal or through a local Web UI with multiple parallel sessions.
- 02Works with any OpenAI-compatible model via user-supplied API keys.
- 03Provides 16 core tools, with other capabilities loaded through a Skill system.
- 04Lets users create Skills from natural language and invoke them with /.
- 05Offers Manual and Auto permission modes for tool approval.
- 06Installs via desktop installer, RubyGems, shell script, or Docker.
Use cases
- Using an AI assistant for everyday coding tasks such as scaffolding a project or exploring an unfamiliar codebase.
- Building reusable Skills that package repeated workflows for a team.
- Running an agent against your choice of models to control API spend.
- Operating an agent on a shared Linux host through the Web UI with an access key.
Quick start
- 1
Install the gem (requires Ruby 3.1 or newer)
gem install openclacky - 2
Start the interactive agent in your project directory
openclacky - 3
In the session, run /config to set your API Key, Model, and Base URL
Things to know
- Requires Ruby 3.1 or newer for the gem install.
- You must supply your own API key for an OpenAI-compatible model (BYOK).
- Browser automation may be limited when running in Docker on macOS or Windows.
In collections
Topics
FAQ
- What is openclacky?
- An open-source AI agent in Ruby that runs coding and general tasks with a minimal tool set, using any OpenAI-compatible model. OpenClacky is an open-source AI agent that can be used from a terminal or a local Web UI for coding, writing, and research sessions. It works with any OpenAI-compatible API, with built-in setup for Claude, GPT, DeepSeek, Kimi, MiniMax, and OpenRouter, and users supply their own API keys.
- Is openclacky free to use? What license does it have?
- openclacky is open source under the MIT license. Check the license terms before using it in commercial products.
- Is openclacky still maintained?
- Actively maintained. The last commit to openclacky was on Nov 7, 2026.
- What can I use openclacky for?
- Using an AI assistant for everyday coding tasks such as scaffolding a project or exploring an unfamiliar codebase. Building reusable Skills that package repeated workflows for a team. Running an agent against your choice of models to control API spend. Operating an agent on a shared Linux host through the Web UI with an access key.
- How do I install openclacky?
- Install the gem (requires Ruby 3.1 or newer): gem install openclacky Start the interactive agent in your project directory: openclacky In the session, run /config to set your API Key, Model, and Base URL
- Where can I get openclacky?
- The source code and installation instructions are on GitHub: https://github.com/clacky-ai/openclacky
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A terminal coding agent written in Rust, scriptable and reloadable through Lua plugins, with a custom renderer and Vim-style editing.
An open source AI coding agent that runs in your terminal, with a desktop app and two built-in agents for building and planning changes.
An open-source AI agent from Google that brings Gemini into your terminal to query, edit, and automate code.
An open-source coding-agent CLI that runs in the terminal and works with cloud APIs and local model backends like Ollama.
An AI coding agent for the terminal that reads and edits code, runs shell commands, and works with Kimi and compatible models.