pilot-shell
The professional development environment for Claude Code. Claude Code is powerful. Pilot Shell makes it reliable. Start a task, grab a coffee, come back to production-grade code. Tests enforced. Context preserved. Quality automated.
claude mcp add --transport stdio maxritter-pilot-shell node path/to/server.js \ --env PILOT_SHELL_CONFIG="path/to/config (optional)" \ --env CLAUDE_CODE_API_KEY="your Claude Code API key (if required by the server)"
How to use
Pilot Shell is a professional development environment designed to bring structure and quality to Claude Code-driven workflows. This MCP server exposes the Pilot Shell tooling as an MCP service so you can drive tasks, enforce tests, and preserve context across sessions within your project. With the server configured, you can invoke the Pilot Shell commands, run spec-driven tasks, and leverage built-in linting, formatting, and type-checking hooks that are integrated into the development workflow. The MCP server acts as your centralized interface to start tasks, fetch results, and ensure your codebase stays aligned with your defined rules and conventions.
Once running, you can use the MCP client tooling to submit work items (like a new feature spec or a bugfix task) and receive a structured plan, implementation steps, and verification results. The system emphasizes enforced testing (TDD-loop), context preservation across sessions, and automatic quality gates. If you already have Claude Code installed in your project, Pilot Shell augments it by providing a guardrail layer that ensures every edit passes linting, formatting, and type checks before being considered complete.
How to install
Prerequisites:
- Node.js (recommended LTS) installed on your system
- curl or wget for script-based installation
- Access to Claude Code (as required by Pilot Shell)
Installation steps:
- Install the Pilot Shell tooling globally or set up your environment as described by the project:
curl -fsSL https://raw.githubusercontent.com/maxritter/pilot-shell/main/install.sh | bash
- Ensure Node.js is available in your PATH. You can verify with:
node -v
npm -v
- If you are wiring up the MCP server manually, place your server entry (e.g., server.js) at a known path and start it using Node:
node /path/to/server.js
-
Create or update your MCP configuration to reference the server name and command. For example, in your MCP config JSON, reference the server as shown in the documentation (see mcp_config in the JSON example).
-
Start your MCP client or orchestration as appropriate for your environment to begin interacting with the Pilot Shell MCP server.
Additional notes
Tips and notes:
- The Pilot Shell MCP server integrates with your project to enforce TDD, linting, and formatting on every edit. Ensure your repository includes the necessary hooks and rules that Pilot Shell relies on.
- If you encounter environment-related issues, verify that CLAUDE_CODE_API_KEY (or equivalent) is correctly set in the server environment.
- For cross-platform use (macOS, Linux, Windows via WSL2), ensure your installed Terminal or shell supports the integration features used by Pilot Shell.
- When configuring MCP, use a stable server path (path/to/server.js) and avoid moving the server binary once running to prevent broken references.
- Review the project’s installation script and prerequisites to confirm compatibility with your OS and shell.
- If the project publishes an npm package, you can switch to an npx or npm-based startup workflow by adapting the mcp_config to your package name and entrypoint.
Related MCP Servers
autosteer
Desktop app for multi-workspace Claude Code management
fast-filesystem
A high-performance Model Context Protocol (MCP) server that provides secure filesystem access for Claude and other AI assistants.
mcp-domain-availability
A Model Context Protocol (MCP) server that enables Claude Desktop to check domain availability across 50+ TLDs. Features DNS/WHOIS verification, bulk checking, and smart suggestions. Zero-clone installation via uvx.
mcp-task-orchestrator
A Model Context Protocol server that provides task orchestration capabilities for AI assistants
context-lens
Semantic search knowledge base for MCP-enabled AI assistants. Index local files or GitHub repos, query with natural language. Built on LanceDB vector storage. Works with Claude Desktop, Cursor, and other MCP clients.
project
Intent-based MCP server for project documentation search. Maps natural language queries to the right sources automatically—no configuration needed. The standard for AI agent documentation search.