# openclaude-multi-llm

Lets Claude talk to models that expose an OpenAI Chat Completions endpoint while keeping Bash, Grep, MCP and the rest of the toolset.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 716 installs · ★ 88
- Author: reason-machines
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/openclaude-multi-llm/SKILL.md
- Updated: 2026-10-09
- HTML: https://affpapa.org/en/nevibecode/skills/openclaude-multi-llm

## What it does

OpenClaude is a Claude fork that sends every model call through a shim, `openaiShim.ts`. The shim stands in for the Anthropic SDK interface, so the rest of the codebase never learns it is talking to someone else. It maps Anthropic content blocks onto the OpenAI `messages` array, turns `tool_use` and `tool_result` into `function_calls` and `tool`, converts OpenAI streaming chunks into Anthropic stream events, and moves system prompt arrays into the `system` role.

The toolset survives the swap: Bash, FileRead, FileWrite, FileEdit, Glob, Grep, WebFetch, Agent, MCP, Tasks, LSP, NotebookEdit. Provider choice lives in environment variables. Skip `CLAUDE_CODE_USE_OPENAI=1` and the shim stays off, so calls go to the Anthropic SDK instead. Cloud providers want `OPENAI_API_KEY`; local Ollama and LM Studio do not. When both are set, `OPENAI_MODEL` wins over `ANTHROPIC_MODEL`.

Install globally with npm: `npm install -g @gitlawb/openclaude`, which gives you the `openclaude` command. Building from source needs Bun. Provider settings can be saved to `.openclaude-profile.json`, so you stop re-exporting variables each session. `bun run doctor:runtime` checks the provider setup before you start.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill openclaude-multi-llm -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/openclaude-multi-llm/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/openclaude-multi-llm/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill openclaude-multi-llm -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/openclaude-multi-llm/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill openclaude-multi-llm -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/openclaude-multi-llm/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install reason-machines/trending-skills openclaude-multi-llm
```

2. **Manually: put the files in place**
- File: `.github/skills/openclaude-multi-llm/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill openclaude-multi-llm -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/reason-machines/trending-skills --path skills/openclaude-multi-llm
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/openclaude-multi-llm/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill openclaude-multi-llm -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/openclaude-multi-llm/SKILL.md

## When to use it
- run Claude tasks on GPT-4o or DeepSeek
- point Claude at a local Ollama instance
- switch models with one environment variable
- save provider settings to a profile once
- check provider reachability before launching

## Example prompts
- Set up an Ollama profile with llama3.1:8b and launch openclaude
- Switch to DeepSeek through `OPENAI_BASE_URL` and list which tools are still available
- Run `bun run doctor:runtime` and tell me what is wrong with the provider

## Similar
- [setup-matt-pocock-skills](https://affpapa.org/en/nevibecode/skills/setup-matt-pocock-skills.md) — skill · 1052851 installs
- [ask-matt](https://affpapa.org/en/nevibecode/skills/ask-matt.md) — skill · 766587 installs
- [to-tickets](https://affpapa.org/en/nevibecode/skills/to-tickets.md) — skill · 711230 installs
- [writing-plans](https://affpapa.org/en/nevibecode/skills/writing-plans.md) — skill · 282859 installs
- [caveman-review](https://affpapa.org/en/nevibecode/skills/caveman-review.md) — skill · 411800 installs
- [caveman-compress](https://affpapa.org/en/nevibecode/skills/caveman-compress.md) — skill · 411500 installs

From the same set: [reason-machines/trending-skills](https://affpapa.org/en/nevibecode/by/reason-machines/trending-skills.md) — 139 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md