A powerful integration for managing Google Jules AI coding agents to autonomously execute software engineering tasks via REST API.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install jules-api
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install jules-api using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Jules API Skill?
The Jules API skill enables developers to harness the power of Google Jules, an autonomous AI coding agent designed to handle complex engineering tasks. By integrating this skill through Openclaw Skills, users can programmatically initiate coding sessions that interact directly with GitHub repositories to write code, fix bugs, and add tests. This skill acts as a bridge between high-level natural language instructions and technical execution, providing a robust interface for modern AI-driven development workflows.
Built on a RESTful architecture, this skill allows for deep customization of the agent's behavior. Whether you need to automate routine maintenance or trigger sophisticated code transformations from your own internal tools, the Jules integration provides the necessary endpoints to manage the entire lifecycle of an AI coding session, from planning and execution to the final pull request generation.
Jules API Skill Use Cases
- Automating the generation of unit tests for existing modules across various repositories.
- Triggering autonomous bug-fixing sessions directly from error monitoring alerts.
- Managing bulk code refactoring tasks programmatically through a centralized dashboard.
- Integrating AI coding capabilities into custom CI/CD pipelines for automated PR creation.
How Jules API Skill Works
- The skill authenticates with the Google Jules backend using a specialized API key passed in the request header.
- Users query for connected GitHub repositories (Sources) to define the execution context for the AI.
- A new session is initialized with a specific task prompt, source branch, and optional automation settings.
- The agent moves through logical states, including planning, awaiting approval, and active execution.
- Developers monitor progress through the activities endpoint, which provides real-time event logging and artifacts like git patches.
- Upon completion, the session provides the final output, such as a created pull request or a set of code changes.
Jules API Skill Setup
To get started with this skill, you must first obtain an API key from the Jules settings page. Ensure your GitHub repositories are connected via the Jules web UI. Set your environment variable as follows:
export JULES_API_KEY='your_api_key_here'
You can then verify your connection by listing your available sources:
curl -s -H "x-goog-api-key: $JULES_API_KEY" "https://jules.googleapis.com/v1alpha/sources"
Jules API Skill Data Schema & Taxonomy
The Jules API skill utilizes a structured metadata system to track the lifecycle of coding tasks. Key data structures include:
| Object | Description |
|---|---|
| Session | Represents a unique coding task, including its state (QUEUED, PLANNING, COMPLETED) and source context. |
| Activity | A granular event log for a session, containing one of several event types like planGenerated or sessionCompleted. |
| Artifact | Technical outputs included in activities, such as ChangeSet (git patches) or BashOutput (command results). |
| Source | Metadata regarding connected GitHub repositories, including branch information and repository identifiers. |
Jules API Skill Advanced Features
- Automation Mode: Support for AUTO_CREATE_PR to bypass manual pull request creation after task completion.
- Interactive Feedback: Use the sendMessage endpoint to provide additional guidance to the agent during AWAITING_USER_FEEDBACK states.
- Granular Polling: Filter session activities by timestamp to efficiently monitor progress without redundant data retrieval.
- Plan Approval Workflows: Explicit requirePlanApproval setting to ensure human-in-the-loop oversight for critical code changes.
- Multi-format Artifacts: Support for retrieving code diffs, command line outputs, and binary media generated during the coding session.
SKILL.md
Loading
Related Openclaw Skills
AI Market Intelligence Agent
An automated intelligence tool that aggregates cross-platform data to generate structured market reports on trends, competitors, and risks.
Web Automation Service
A robust browser automation tool designed to handle form submissions, web scraping, and scheduled monitoring tasks via AI-driven workflows.
Revenue Tracker
A lightweight CLI-based system for logging daily income and generating detailed financial performance reports.
Data Scraping Service
A high-performance automated tool for web scraping, API extraction, and data cleaning.
CFO Financial Operations
A comprehensive AI financial operations skill designed to manage unit economics, runway calculations, and revenue forecasting for startups.
CMO Content Marketing Skill
An AI agent skill designed to manage content strategy, multi-modal brand development, and LinkedIn demand generation with a focus on revenue impact.
METADATA
Featured*
