Basalt — AI skill for Claude Code
Turn your Mac into a local LLM server: MLX on Metal, OpenAI + Anthropic API + MCP, pure Swift.
How to install Basalt
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open goodmai/basalt and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Basalt does
Turn your Mac into a local LLM server: MLX on Metal, OpenAI + Anthropic API + MCP, pure Swift. Point Claude Code at it.
Alternatives in AI
- WindsurfAPI — Turn Windsurf / Devin Desktop's 100+ AI models (Claude, GPT, Gemini, DeepSeek, Kimi, GLM, SWE) into OpenAI-, A 3k ★
- Vllm Mlx — High-performance OpenAI and Anthropic compatible LLM inference server for Apple Silicon 1.5k ★
- Claude To ChatGPT — This project converts the API of Anthropic's Claude model to the OpenAI Chat API format 1.3k ★
README
basalt — turn your Mac into a local LLM server. Point Claude Code at it.
Gemma 4, Qwen 3 and any MLX model on your Apple Silicon GPU, behind the **OpenAI API, the Anthropic API and MCP**. One Swift binary. No Python at runtime. No cloud. No API bill.
[English](README.md) · [Русский](README.ru.md)
[](https://github.com/goodmai/basalt/actions/workflows/release.yml) [](https://github.com/goodmai/basalt/actions/workflows/security-audit.yml) [](LICENSE)
[](#requirements) [](Package.swift) [](#install-with-homebrew) [](#benchmarks) [](#benchmarks)
┌──────────────────────────────────────────────────────┐
│ Basalt │
│ │
│ MCP stdio ──┐ │
│ ├──► ModelOrchestratorActor ──► MLX │
│ REST :8080 ─┘ (actor, FIFO) Metal GPU │
└──────────────────────────────────────────────────────┘
Two transports share a single actor instance — **MCP stdio** for IDE integration (Claude Desktop, Cursor) and **REST HTTP** for agent-to-agent workflows and Claude Code backends.
Related Skills
AgentOS
Turn your iPhone into an AI agent host. 165+ native tools, 30 agents, 10 channels, 5 AI providers. Pure Swift
TurboLLM
Run any local LLM engine, auto-tuned to your GPU — polished web UI + OpenAI/Anthropic-compatible API. Point Cl
Wax
Shared Single-file memory layer for all your agents, sub mili-second RAG over text, photo and video on Apple S
Agent Rs
Pure-Rust async runtime for LLM agents. Multi-provider (Anthropic/OpenAI/Ollama), tool-capable end-to-end, str
Yunshu
Fast local LLM / VLM inference engine for Apple Silicon (MLX). OpenAI- and Anthropic-compatible, lossless spec
Superfluid
One serving daemon for local LLM agents: durable sessions, QoS scheduling and OpenAI, Anthropic and Ollama API
Related Agents
LLM Orchestrator
Use this agent for LLM integration work — prompt engineering, multi-provider abstraction (OpenRouter, Gemini,
Bonsai
Runs a task on the local Bonsai 2 27B model (PrismML ternary 27B, MLX, on this Mac) and returns its answer. Us
AI LLM Engineer
Use this agent when the user types /ai or asks for ai/llm engineer work — e.g., build a RAG pipeline over our