# runx

runx lets an agent find, inspect, and run skills through an MCP connection, while recording a verifiable receipt for each execution.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed · ★ 104
- Author: runxhq
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/runxhq/runx
- Updated: 2026-10-22
- HTML: https://affpapa.org/en/nevibecode/mcp/runx

## What it does

runx gives an agent a controlled way to execute skills from a local directory, a URL, or a catalog. The agent can discover a skill, inspect what it needs, connect selected tools, and pass its result into the next step. Each run keeps state and produces a receipt that the CLI can verify.

The MCP server works with Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, and Zed. Execution, state, and receipt signing happen locally; a hosted registry, credential storage, and notary services are separate options. The repository example requires Git and Node.js 20 or newer. It does not require a Runx account, model, or provider key.

For Claude Code, add the server with one command. After cloning the repository, the local example runs with `runx skill ./examples/hello-world --run-id first-run -i message="Hello, Runx." --json`. You can inspect it with `runx history first-run --json` and verify the returned receipt using `runx verify <receipt-id> --json`.

## Requirements
- Remote server: `https://api.runx.ai/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http runx https://api.runx.ai/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add runx --url https://api.runx.ai/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "runx": {
            "url": "https://api.runx.ai/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "runx": {
            "type": "http",
            "url": "https://api.runx.ai/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http runx https://api.runx.ai/mcp
```

## When to use it
- inspect a skill before execution
- pass one skill’s result into the next step
- limit a run to selected tools
- verify the receipt produced by a run

## Example prompts
- Connect runx and find a skill that can examine where users leave our funnel.
- Run the skill in `./examples/hello-world` with run ID `first-run` and the message “Hello, Runx.”.
- Read the history for `first-run`, then verify its receipt.

## Similar
- [REA](https://affpapa.org/en/nevibecode/mcp/rea.md) — MCP server · ★ 386989
- [GitHub](https://affpapa.org/en/nevibecode/mcp/github.md) — MCP server · ★ 33918
- [Netdata](https://affpapa.org/en/nevibecode/mcp/netdata.md) — MCP server · ★ 81115
- [browser-use](https://affpapa.org/en/nevibecode/mcp/browser-use.md) — MCP server · ★ 119160
- [Codebase Memory](https://affpapa.org/en/nevibecode/mcp/codebase-memory.md) — MCP server · ★ 47392
- [mcp-server](https://affpapa.org/en/nevibecode/mcp/mcp-server.md) — MCP server · ★ 43872

---
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