sidekick-cli

MCP.Pizza Chef: geekforbrains

Sidekick-cli is an open-source, agentic CLI client designed for AI-driven development workflows. It supports multiple LLM providers, enabling users to switch models seamlessly within sessions. Featuring MCP protocol support, undo commands, per-project customization, and cost tracking, Sidekick-cli offers a flexible, CLI-first alternative to traditional IDE-based AI tools. It is ideal for developers seeking vendor-agnostic, command-line AI assistance with real-time context and control.

Use This MCP client To

Develop code interactively using AI in the command line Switch between different LLM providers during development sessions Customize AI behavior per project with configuration guides Undo AI-generated changes instantly with /undo command Track token usage and cost per command or session Inject system prompts dynamically to maintain AI context Operate AI workflows without relying on IDEs or vendor lock-in

README

Sidekick (Beta)

Sidekick Demo

Your agentic CLI developer.

Overview

Sidekick is an agentic CLI-based AI tool inspired by Claude Code, Copilot, Windsurf and Cursor. It's meant to be an open source alternative to these tools, providing a similar experience but with the flexibility of using different LLM providers while keeping the agentic workflow.

Sidekick is currently in beta and under active development. I'd love your ideas and feedback.

Features

  • No vendor lock-in. Use whichever LLM provider you prefer.
  • MCP support
  • Use /undo when AI breaks things.
  • Easily switch between models in the same session.
  • JIT-style system prompt injection ensures Sidekick doesn't lose the plot.
  • Per-project guide. Adjust Sidekick's behavior to suit your needs.
  • CLI-first design. Ditch the clunky IDE.
  • Cost and token tracking.
  • Per command or per session confirmation skipping.

Roadmap

  • Tests 😅
  • More LLM providers, including Ollama

Quick Start

Install Sidekick.

pip install sidekick-cli

On first run, you'll be asked to configure your LLM providers.

sidekick

Configuration

After initial setup, Sidekick saves a config file to ~/.config/sidekick.json. You can open and edit this file as needed. Future updates will make editing easier directly from within Sidekick.

MCP Support

Sidekick supports Model Context Protocol (MCP) servers. You can configure MCP servers in your ~/.config/sidekick.json file:

{
  "mcpServers": {
    "fetch": {
      "command": "uvx",
      "args": ["mcp-server-fetch"]
    },
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}

MCP servers extend the capabilities of your AI assistant, allowing it to interact with additional tools and data sources. Learn more about MCP at modelcontextprotocol.io.

Available Commands

  • /help - Show available commands
  • /yolo - Toggle "yolo" mode (skip tool confirmations)
  • /clear - Clear message history
  • /compact - Summarize message history and clear old messages
  • /model - List available models
  • /model <num> - Switch to a specific model (by index)
  • /undo - Undo most recent changes
  • /dump - Show current message history (for debugging)
  • exit - Exit the application

Customization

Sidekick supports the use of a "guide". This is a SIDEKICK.md file in the project root that contains instructions for Sidekick. Helpful for specifying tech stack, project structure, development preferences etc.

Telemetry

Sidekick uses Sentry for error tracking and usage analytics. You can disable this by starting with the --no-telemetry flag.

sidekick --no-telemetry

Installation

Using pip

pip install sidekick-cli

From Source

  1. Clone the repository
  2. Install dependencies: pip install . (or pip install -e . for development)

Development

# Install development dependencies
make install

# Run linting
make lint

# Run tests
make test

License

MIT

sidekick-cli FAQ

How do I switch between different LLM providers in Sidekick-cli?
You can easily switch between supported LLM providers within the same session using built-in commands, allowing flexible AI model usage.
What does MCP support mean in Sidekick-cli?
MCP support means Sidekick-cli can integrate with the Model Context Protocol to provide structured, real-time context and tool access to LLMs, enhancing AI interactions.
How can I undo AI-generated changes?
Use the /undo command to revert any AI-generated modifications instantly, ensuring safe and controlled development.
Is Sidekick-cli limited to a specific IDE or environment?
No, Sidekick-cli is CLI-first and works independently of any IDE, providing a lightweight and flexible AI development experience.
Can I customize Sidekick-cli behavior for different projects?
Yes, Sidekick-cli supports per-project guides to tailor AI behavior and responses to your specific needs.
How does Sidekick-cli help manage costs?
It tracks token usage and associated costs per command or session, helping you monitor and control your AI expenses.
Is Sidekick-cli open source?
Yes, Sidekick-cli is an open-source project, encouraging community contributions and transparency.
What LLM providers does Sidekick-cli support?
Sidekick-cli supports multiple LLM providers and plans to add more, including Ollama, allowing vendor flexibility alongside providers like OpenAI, Anthropic, and Claude.