# multi-llm-mcp-server

The multi-llm-mcp-server skill connects Claude Code with Codex CLI and multiple language models. Use it to compare parallel model responses or hand code analysis to Codex.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 107 installs · ★ 9 repo stars
- Author: reason-machines
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/reason-machines/mcp-skills/tree/HEAD/skills/multi-llm-mcp-server/SKILL.md
- Updated: 2026-10-09
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/multi-llm-mcp-server

## What it does

The skill guides Claude Code in sending a prompt to one model or several at once, then comparing the results. Multiple models can inspect the same code with a shared instruction. Sessions let you continue a conversation with a chosen model, review active sessions, or clear them.

The MCP server requires Python 3.10+, the `fastmcp` and `openai` packages, and API keys for the providers you plan to use. Its `ask_codex` and `wait_codex` tools connect to Codex CLI. In `read-only` mode Codex can inspect files without changing them; `workspace-write` allows edits inside the workspace. Codex CLI must be installed, authenticated, and available on `PATH` for those tools to run.

The skill covers server setup and tools for checking configured providers and waiting on longer jobs. Claude requires a manual change to `LLM_MIX.py` and an `ANTHROPIC_API_KEY`; until enabled, that provider is not available through the server.

## Install

### Claude Code (native platform)

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

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

### Codex

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

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

### Cursor

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

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

### GitHub Copilot

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

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

### Gemini CLI

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

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

### Windsurf (Devin Desktop)

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

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add reason-machines/mcp-skills --skill multi-llm-mcp-server -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/mcp-skills/tree/HEAD/skills/multi-llm-mcp-server/SKILL.md

## When to use it
- Compare several models' answers to one prompt
- Ask multiple models to review the same code
- Have Codex CLI inspect a project's structure
- Delegate workspace file edits to Codex CLI
- Check which model providers are configured

## Example prompts
- Compare GPT, DeepSeek, and Qwen on whether this project should use a monorepo or separate repositories.
- Review this code for security issues using multiple models.
- Ask Codex CLI to find every TODO comment in the project without changing files.

## Similar
- [openclaude-multi-llm](https://affpapa.org/en/nevibecode/skills/openclaude-multi-llm.md) — skill · 725 installs
- [multi-tenant-llm-hosting](https://affpapa.org/en/nevibecode/skills/multi-tenant-llm-hosting.md) — skill · 344 installs
- [auto-review-loop-llm](https://affpapa.org/en/nevibecode/skills/auto-review-loop-llm.md) — skill · 550 installs
- [LLM Council](https://affpapa.org/en/nevibecode/skills/shuntacurosu-llm-council.md) — skill · 5 installs
- [multi-llm](https://affpapa.org/en/nevibecode/plugins/multi-llm.md) — plugin · ★ 1596 repo stars
- [om-fix-issue-multi](https://affpapa.org/en/nevibecode/skills/om-fix-issue-multi.md) — skill · 500 installs

From the same set: [reason-machines/mcp-skills](https://affpapa.org/en/nevibecode/by/reason-machines/mcp-skills.md) — 98 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
