runtool

The bridge between human and AI tooling

Define functions in a Runfile. Your AI agent discovers and executes them via the built-in MCP server. You run them from the terminal too with instant startup and tab completion. Shell, Python, Node—whatever fits the task.

# @desc Search the codebase for a pattern
# @shell python
search(pattern: str) {
    import sys, os, re
    for root, _, files in os.walk('.'):
        for f in files:
            path = os.path.join(root, f)
            try:
                for i, line in enumerate(open(path), 1):
                    if re.search(sys.argv[1], line):
                        print(f"{path}:{i}: {line.rstrip()}")
            except: pass
}

Humans run it from the terminal:

$ run search "TODO"

AI agents discover and execute it automatically via the Model Context Protocol (MCP).

Note: runtool is an alternative package name for run. Both install the same run binary and provide identical functionality.

Installation

cargo install runtool

This installs the run binary to your cargo bin directory.

Recommended Installation

macOS/Linux (Homebrew)

brew tap nihilok/tap
brew install runtool

Windows (Scoop)

scoop bucket add nihilok https://github.com/nihilok/scoop-bucket
scoop install runtool

Quick Start

Create a Runfile in your project:

# @desc Deploy to an environment
deploy(env: str, version = "latest") {
    echo "Deploying $version to $env..."
    ./scripts/deploy.sh $env $version
}

# @desc Run the development server
dev() {
    cargo run --release
}

Run from terminal:

$ run deploy staging
$ run dev

Enable AI integration:

$ run --serve-mcp

Full Documentation

See the run crate documentation for:

• AI Agent Integration (MCP)
• Complete syntax guide
• Function composition
• Polyglot scripts (Python, Node, Ruby, etc.)
• Tab completions
• Platform-specific functions

Source Code

Both run and runtool are maintained in the same repository: https://github.com/nihilok/run