Enables automated code modification and repository management through Jules, supporting repository-specific sessions and the retrieval of pull request URLs for completed tasks.
Provides tools to interact with Jules, Google's AI coding agent, allowing for the management of sessions, approval of execution plans, and direct communication with the agent.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Jules API MCP Start a new session to fix the memory leak in owner/repo"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP server for Jules — Google's AI coding agent. Manage Jules sessions from any MCP-compatible client: Claude, Gemini CLI, Cursor, Zed, VS Code, and more.
Tool | Description |
| List all Jules sessions |
| Get session details and current state |
| Create a new Jules session |
| Approve Jules's plan to proceed with execution |
| Send feedback or instructions to Jules |
| Full activity timeline for a session |
| Jules's latest outbound message |
| Get the PR URL from a completed session |
Related MCP server: Jules MCP Server
Node.js 18+
A Jules API key (JULES_API_KEY)
git clone https://github.com/GreyC/jules-api-mcp
cd jules-api-mcp
npm install
npm run buildAdd to your MCP client config (e.g. ~/.claude/claude_desktop_config.json or ~/.gemini/settings.json):
{
"mcpServers": {
"jules": {
"command": "node",
"args": ["/path/to/jules-api-mcp/dist/index.js"],
"env": {
"JULES_API_KEY": "<your-key>"
}
}
}
}List all sessions:
list_sessions()Create a session with auto PR:
create_session(
repo="owner/repo",
prompt="Refactor the login function in auth.py to use async/await",
automation_mode="AUTO_CREATE_PR"
)Approve a plan:
approve_plan(session_id="<ID>")Send feedback:
send_message(session_id="<ID>", message="Looks good, please create the PR.")State | Meaning |
| Jules is actively working |
| Jules generated a plan, waiting for approval |
| Jules has a question mid-execution |
| Jules finished |
| Unrecoverable error |
jules-skill — Agent skill with workflow guidance, decision heuristics, and prompt templates for autonomous Jules management
MIT
This server cannot be deployed
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
MCP server for building and testing AI agents with multi-model experimentation and insights.
MCP server for AI agents to plan, verify, and deploy Cloudflare-native apps.
