Google Jules
Google Jules is an AI coding agent developed by Google to assist developers with software development tasks. It understands codebase context, autonomously plans and executes programming tasks, supports integration with external tools via API calls, maintains a memory system to preserve task context coherence, and provides a command-line interface (CLI) for developers. It is applicable to software engineering scenarios such as code generation, debugging, and refactoring.
Core Facts
Google Jules is an asynchronous AI coding agent launched by Google in late 2024, based on the Gemini model.
OpenAI's Codex and Google's Jules also possess background agent capabilities similar to Cursor's Background Agent
Google Jules is an AI coding agent for developers launched by Google, capable of autonomously completing tasks such as code modifications and bug fixes.
Timeline (last 90 days)
Claude.md is loaded at the start of each session; the Agents.md format originated from collaboration among OpenAI Codex, AMP, Google Jules, Cursor, Factory, and others, and is now maintained by the Agentic AI Foundation.
When using Jules, developers' code is uploaded and run on Google's infrastructure during task execution, and task triggering and monitoring can only be done through the official interface.
Google Jules runs in a sandbox environment on Google's cloud, with each task executed inside an isolated virtual machine.
jules-app is a self-hosted workbench project built for the Google Jules AI agent.
Google Jules belongs to the same "autonomous coding Agent" track as products such as GitHub Copilot Workspace and Devin.
There is a repository on GitHub named warrenet/Jules-ai-code-related-, described as "Coding from Google Jules".
Google Jules entered public beta at the end of 2024.
Jules is currently still in the waitlist stage.
Jules's "asynchronous, reviewable" working model is better suited for handling independent tasks with clear boundaries (such as fixing Issues or refactoring modules) rather than real-time pair programming.
Jules runs in a sandboxed cloud VM, can clone a user's GitHub repository and execute changes within it, and ultimately presents the results in the form of a PR for developer review.
12 more timeline events
All Facts (20)
Google Jules is an asynchronous AI coding agent launched by Google in late 2024, based on the Gemini model.
75%VerifiedOpenAI's Codex and Google's Jules also possess background agent capabilities similar to Cursor's Background Agent
65%UnverifiedJules的工作模式与GitHub深度集成，能够自动克隆仓库、分析代码库、制定修改计划、编写代码，最终以Pull Request的形式提交结果
85%Partially VerifiedGoogle Jules is an AI coding agent for developers launched by Google, capable of autonomously completing tasks such as code modifications and bug fixes.
75%UnverifiedJules belongs to the category of AI coding agents, and is designed to autonomously complete tasks such as code modifications and bug fixes under developer authorization.
60%Unverifiedgoogle-jules-mcp is an MCP server project that acts as a secure bridge connecting AI clients to the Google Jules API.
60%UnverifiedGoogle Jules is backed by the Gemini model's capabilities and provides a dedicated Agentic API for programmatic access.
60%UnverifiedClaude.md is loaded at the start of each session; the Agents.md format originated from collaboration among OpenAI Codex, AMP, Google Jules, Cursor, Factory, and others, and is now maintained by the Agentic AI Foundation.
50%UnverifiedWhen using Jules, developers' code is uploaded and run on Google's infrastructure during task execution, and task triggering and monitoring can only be done through the official interface.
50%Unverifiedjules-app is a self-hosted workbench project built for the Google Jules AI agent.
50%UnverifiedGoogle Jules runs in a sandbox environment on Google's cloud, with each task executed inside an isolated virtual machine.
50%UnverifiedJules runs in a sandboxed cloud VM, can clone a user's GitHub repository and execute changes within it, and ultimately presents the results in the form of a PR for developer review.
50%UnverifiedJules's "asynchronous, reviewable" working model is better suited for handling independent tasks with clear boundaries (such as fixing Issues or refactoring modules) rather than real-time pair programming.
50%UnverifiedJules is currently still in the waitlist stage.
50%UnverifiedGoogle Jules belongs to the same "autonomous coding Agent" track as products such as GitHub Copilot Workspace and Devin.
50%UnverifiedJules clones a code repository, formulates a plan, executes modifications in a sandboxed cloud environment on Google's infrastructure, and ultimately returns the results in the form of a Pull Request.
50%UnverifiedThe autonomous-agents project was inspired by Google Jules — an asynchronous coding agent launched by Google.
50%UnverifiedJules can autonomously pull code in the background, analyze context, generate changes, and submit results in the form of a Pull Request, without requiring continuous human intervention.
50%Unverifiedautonomous-agents takes a differentiated approach from Jules — Jules leans toward being a general-purpose end-to-end coding agent, while autonomous-agents emphasizes multi-agent collaboration and domain specialization.
50%Unverifiedgoogle-jules-mcp primarily provides support for three categories of Jules agent capabilities: code generation, plan approval, and session management.
50%