Metadata-Version: 2.4
Name: agent-client-kernel
Version: 0.2.0
Summary: A Jupyter Kernel for Agent Client Protocol using MetaKernel
Project-URL: Homepage, https://github.com/jimwhite/agent-client-kernel
Project-URL: Repository, https://github.com/jimwhite/agent-client-kernel
Author-email: Jim White <jim@jimwhite.net>
License-Expression: BSD-3-Clause
License-File: LICENSE
Keywords: ACP,IPython,Jupyter,agent,kernel
Classifier: Development Status :: 3 - Alpha
Classifier: Framework :: IPython
Classifier: Framework :: Jupyter
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.10
Requires-Dist: agent-client-protocol>=0.4.0
Requires-Dist: ipykernel>=7.0
Requires-Dist: jupyter-client>=8.5
Requires-Dist: nest-asyncio>=1.6.0
Description-Content-Type: text/markdown

# agent-client-kernel

A Jupyter Kernel for Zed's Agent Client Protocol (ACP) https://agentclientprotocol.com/ .

This kernel allows you to interact with external ACP agents directly from Jupyter notebooks. It acts as an ACP client that connects to coding agents like [Codex](https://github.com/zed-industries/codex-acp), providing a seamless notebook interface for AI-powered coding assistance.

## About

This project implements a Jupyter kernel that serves as a client for coding agents via the Agent Client Protocol. The implementation uses [MetaKernel](https://github.com/Calysto/metakernel) as the base class, which provides built-in magics, shell commands, and other useful features.

The kernel spawns and communicates with external ACP agents (such as codex-acp) via stdio, allowing you to interact with AI coding assistants directly from your notebook.

## Features

- **ACP Client Implementation**: Full client-side ACP protocol support
- **External Agent Integration**: Connects to any ACP-compatible agent
- **Codex Support**: Works with OpenAI Codex out of the box
- **Based on MetaKernel**: Built-in magics (help, shell, file operations, etc.)
- **Configurable**: Easily switch between different agents via environment variables
- **Compatible with JupyterLab and Jupyter Notebook**

## Installation

Just use the Dockerfile by opening the devcontainer in GitHub Codespaces or locally.  
After JupyterLab starts, open a Terminal and run `codex`.  Follow prompts for API key and authorization then `/quit`.

Alternatively, install the package, agent, and kernel:

```bash
pip install --upgrade uv jupyter-mcp-tools
git submodule update --init --recursive
pip install -e .
python -m agentclientkernel install --user
```

## Configuration

The kernel spawns a subprocess to run the agent which needs installation and ACP adapter.

### Using Codex (Default)

1. Install codex-acp:
   ```bash
   npm install -g @openai/codex@latest @zed-industries/codex-acp@latest
   ```

2. Set your OpenAI API key:  This can be onitted and `codex` will prompt for this.
   ```bash
   export OPENAI_API_KEY=sk-...
   ```

3. Authorize Codex
   ```bash
   codex
   ```

   It will prompt you through authentication and permission to run stuff.

   I think only API auth works in Codespaces because OAuth tries to redirect thru localhost.

   This is the error you get when trying to chat with the agent then you probably missed this step.
   ```
   Error: Authentication required

   Make sure the ACP agent is configured correctly.
   Current agent: codex-acp
   ```

4. Start Jupyter and use the "Agent Client Protocol" kernel
   ```bash
   start-noteboook.py
   ```

### Using Other Agents

Set environment variables to configure a different agent:

```bash
export ACP_AGENT_COMMAND=path/to/your/agent
export ACP_AGENT_ARGS="--arg1 --arg2"
```

Then start Jupyter normally. The kernel will use your configured agent.

## Usage

After installation, create a new notebook and select "Agent Client Protocol" as the kernel.

Type your prompts in cells and execute them to interact with the agent:

```
Create a Python function to calculate fibonacci numbers
```

The agent will respond with code, explanations, and can help with:
- Writing code
- Debugging
- Code review
- Refactoring
- Documentation
- And more!

## Add a Jupyter MCP Service 

Adding a Jupyter MCP service for accessing and editing notebooks and cells.
The Dockerfile installed the datalayer/jupyter-mcp-server https://github.com/datalayer/jupyter-mcp-server .
To add it to the agent's MCP configuration: 
   ```
   %agent mcp add jupyter uvx jupyter-mcp-server@latest
   ```
See examples/jupyter-mcp.ipynb.

### Magic Commands

The kernel provides a unified `%agent` magic command for all configuration and session management:

**MCP Server Configuration:**
- `%agent mcp add NAME COMMAND [ARGS...]` - Add an MCP server
- `%agent mcp list` - List configured MCP servers
- `%agent mcp remove NAME` - Remove an MCP server
- `%agent mcp clear` - Remove all MCP servers

**Permission Configuration:**
- `%agent permissions [auto|manual|deny]` - Set permission mode
- `%agent permissions list` - View permission request history

**Session Management:**
- `%agent session new [CWD]` - Create a new session
- `%agent session info` - Show current session information
- `%agent session restart` - Restart the current session

**Agent Configuration:**
- `%agent config [COMMAND [ARGS...]]` - Configure the agent command
- `%agent env [KEY=VALUE]` - Set agent environment variables

Use `%agent` without arguments to see all available subcommands.
Use `%agent?` for detailed help on the magic command.

See the example notebooks in `examples/` for demonstrations:
- `basic_usage.ipynb` - Basic agent interaction
- `configuration_demo.ipynb` - Configuration and session management

## Requirements

- Python >= 3.10
- ipykernel >= 4.0
- jupyter-client >= 4.0  
- agent-client-protocol >= 0.4.0
- metakernel >= 0.30.0
- An ACP-compatible agent (e.g., codex-acp)

## Uninstallation

```bash
jupyter kernelspec remove agentclient
pip uninstall agentclientkernel
```

## License

BSD 3-Clause License
