Jules MCP Server
Enables LLM applications to interact with Google's Jules AI coding assistant to manage repositories, coding sessions, and pull requests. It allows users to programmatically create tasks, approve plans, and communicate with the assistant during active coding sessions.
README
Jules MCP Server
A Model Context Protocol (MCP) server for the Google Jules API. This server enables LLM applications to interact with Jules - Google's AI coding assistant - to create sessions, send messages, and manage coding tasks programmatically.
Features
- List Sources: Browse available GitHub repositories connected to Jules
- Create Sessions: Start new coding tasks with Jules
- Send Messages: Communicate with Jules during active sessions
- Approve Plans: Approve Jules's proposed plans before execution
- List Activities: View the work history and conversation for a session
- Create Pull Requests: Convenience tool to create sessions that result in PRs
Prerequisites
- Python 3.10+
- A Jules API key (generate one from the Jules Settings page)
- GitHub repositories installed in Jules
Installation
Using uv (Recommended)
# Clone the repository
git clone https://github.com/yourusername/jules-mcp.git
cd jules-mcp
# Install dependencies with uv
uv sync
Using pip
# Clone the repository
git clone https://github.com/yourusername/jules-mcp.git
cd jules-mcp
# Create virtual environment
python -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
# Install dependencies
pip install fastmcp httpx pydantic
Configuration
Set the JULES_API_KEY environment variable with your API key:
export JULES_API_KEY="your-api-key-here"
Usage
Running Standalone
# With uv
uv run python main.py
# With pip/venv
python main.py
MCP Configuration
Add this server to your MCP client configuration. Below are examples for different setups.
Claude Desktop / Claude Code
Add to your mcp.json or claude_desktop_config.json:
{
"mcpServers": {
"jules": {
"command": "uv",
"args": [
"--directory",
"/absolute/path/to/jules-mcp",
"run",
"python",
"main.py"
],
"env": {
"JULES_API_KEY": "your-jules-api-key"
}
}
}
}
Alternative (without uv)
{
"mcpServers": {
"jules": {
"command": "/absolute/path/to/jules-mcp/.venv/bin/python",
"args": [
"/absolute/path/to/jules-mcp/main.py"
],
"env": {
"JULES_API_KEY": "your-jules-api-key"
}
}
}
}
Available Tools
list_sources
List available GitHub repositories that Jules can work with.
Parameters:
page_size(int, optional): Number of sources to return (1-100, default 30)page_token(str, optional): Pagination token
get_source
Get details about a specific source repository.
Parameters:
source_name(str, required): Resource name (e.g., "sources/github/owner/repo")
list_sessions
List Jules sessions (coding tasks).
Parameters:
page_size(int, optional): Number of sessions to return (1-100, default 30)page_token(str, optional): Pagination tokenactive_only(bool, optional): Filter to only show active sessions
get_session
Get details about a specific session.
Parameters:
session_name(str, required): Resource name (e.g., "sessions/abc123")
create_session
Create a new Jules session to work on a coding task.
Parameters:
prompt(str, required): The coding task descriptionsource(str, required): Resource name of the repositorybranch(str, optional): Branch to usetitle(str, optional): Session titlerequire_plan_approval(bool, optional): Wait for plan approval before executing
send_message
Send a follow-up message to an active session.
Parameters:
session_name(str, required): Resource name of the sessionmessage(str, required): Message to send
approve_plan
Approve Jules's plan for a session in AWAITING_PLAN_APPROVAL state.
Parameters:
session_name(str, required): Resource name of the session
list_activities
List activities (work history) for a session.
Parameters:
session_name(str, required): Resource name of the sessionpage_size(int, optional): Number of activities to return (1-100, default 50)page_token(str, optional): Pagination token
create_pull_request
Create a session that will result in a pull request.
Parameters:
prompt(str, required): Description of changes to makesource(str, required): Resource name of the repositorybranch(str, optional): Base branch for the PRtitle(str, optional): Title for the session/PR
Session States
Sessions progress through these states:
QUEUED- Session is waiting to startPLANNING- Jules is creating a planAWAITING_PLAN_APPROVAL- Waiting for user to approve the planAWAITING_USER_FEEDBACK- Waiting for user inputIN_PROGRESS- Jules is working on the taskPAUSED- Session is pausedFAILED- Session failedCOMPLETED- Session completed successfully
Example Workflow
-
List available repositories:
Use the list_sources tool to see which repos are available -
Create a coding task:
Use create_session with: - prompt: "Add a dark mode toggle to the settings page" - source: "sources/github/myorg/myrepo" -
Monitor progress:
Use get_session to check the session state Use list_activities to see what Jules is doing -
Interact with Jules:
Use send_message if Jules needs clarification Use approve_plan if plan approval is required -
Get the result:
Once session is COMPLETED, check the outputs field for PR URL
API Reference
This server wraps the Google Jules API (v1alpha).
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
E2B
Using MCP to run code via e2b.