maki
An efficient AI coding agent extendable by neovim-like Lua plugins
Maki is a high-performance AI coding agent engineered for minimal token consumption while maintaining excellent developer experience. Built with Rust and ratatui, it uses tree-sitter parsing and sandboxed code execution to optimize context window usage, achieving significant token savings compared to traditional coding agents. The agent supports delegation to sub-agents with adjustable model strengths, extends via Neovim-style Lua plugins, and integrates with MCP servers and skills. It provides a 60 FPS terminal interface with granular permission controls, memory management, and compatibility with over a dozen LLM providers including Anthropic, OpenAI, Google, and local models via Ollama or llama.cpp.
Key Features
Use Cases
- 01Developing software projects with cost-conscious token budgets and limited context windows
- 02Delegating complex coding tasks across multiple specialized sub-agents with varying model capabilities
- 03Extending agent capabilities through custom Lua plugins for project-specific workflows
- 04Running AI-assisted coding entirely locally using Ollama or llama.cpp models
- 05Integrating semantic code search and AST-based refactoring via MCP servers and Lua tools
- 06Operating in security-sensitive environments with granular control over file operations and network requests
Related Agents
View morehermes-agent
The agent that grows with you
agency-agents
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
openinterpreter
A coding agent for open models like Kimi K3
cline
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
maki — FAQ
What is maki and what does it do?+
Maki is an AI coding agent optimized to minimize context token usage while providing a fast, transparent terminal-based interface. It uses tree-sitter parsing and sandboxed code execution to efficiently manage context, and can be extended with Lua plugins similar to Neovim.
How do I install the maki AI agent?+
On Linux/macOS, run the install script from maki.sh which installs to ~/.local/bin. On Windows, use the PowerShell installer or Git Bash script. Alternatively, install via cargo, Nix, or download pre-built binaries from GitHub Releases.
Which AI providers does maki support?+
Maki supports Anthropic, OpenAI, xAI, Google Gemini, GitHub Copilot, Ollama, llama.cpp, Mistral, DeepSeek, OpenRouter, and several others. You can also create custom providers by placing executable scripts in ~/.config/maki/providers/.
What API keys or prerequisites are needed to use maki?+
You need an API key for your chosen provider (e.g., ANTHROPIC_API_KEY, OPENAI_API_KEY, GEMINI_API_KEY). For local models, configure OLLAMA_HOST or LLAMA_CPP_HOST. Some providers like OpenAI and xAI also support OAuth login via the maki auth command.
Can maki integrate with code editors like Zed?+
Yes, maki supports the ACP protocol. Run 'maki acp' or configure your ACP-supporting editor (such as Zed) to use maki as an agent server by specifying the maki command with acp argument in your editor settings.
Is maki free and open source?+
Yes, maki is open source and available on GitHub. The software itself is free, though you'll need API access to your chosen LLM provider, which may have associated costs (or use free local models via Ollama).
How do I install maki?+
Open the source repository on GitHub and follow its README. maki is a agent — MCP Agents Market links you directly to the official repo.
Is maki free?+
maki is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.