Google expands its AI coding agent Jules with a command-line interface and a public API, allowing it to plug into terminals, CI/CD systems, and tools like Slack
Jules in the command line — We're launching Jules Tools … Kathy Korevec / The Keyword : New ways to build with Jules, our AI coding agent Kathy Korevec / Google Developers Blog : Meet Jules Tools: A Command Line Companion for Google's Async Coding Agent Craig Hale / TechRadar : Google's AI coding agent Jules is getting new command line tools Joseph Humphrey / Chrome Unboxed : Google just made its AI coding agent, Jules, way more useful for developers X: Chris Perry / @thechrisperry : Fantastic work from our friends at Jules :) John Ennis / @john_ennis_btc : Huge news! Ramón Medrano Llamas / @rmedranollamas : BOOM Dion Almaer / @dalmaer : A big missing piece is here. I stitch my designs... put them in a stitch directory in my repo and when local this is great to fire off tasks. @lalopenguin : so many wayyysss @brextonpham : Easily the best piece of news all week for me, even above sonnet 4.5 and sora Have been asking for this for months!! Go @julesagent N. Taylor Mullen / @ntaylormullen : Undeniable synergy Gemini CLI 🤝 Jules Local agent power will be meeting remote tasking power. Oh, it's happening. Stay tuned. @skiminok : 🐙 Jules is now in your CLI! - Delegate dozens of SWE tasks daily to your remote partner in the cloud. Multitask. Lock in. Scale. - Integrate from your favorite tools. IDE? Subagent? Reviewer? Your imagination is the limit. Kath Korevec / @simpsoka : npm install -g @google/jules Kath Korevec / @simpsoka : Big day 🦑 Jules / @julesagent : Jules just cd-ed into your terminal. 🤖 Introducing Jules Tools, a new lightweight command line interface that lets you do everything you can do in the browser, available now in your terminal. You can get it via npm install -g @google/jules Get the details in the blog → [video] LinkedIn: Kathy K. : Today we're opening up Jules Tools: a command line interface that brings Jules' coding agent directly into developers' workflows. … Shivani Poddar : Do you know that feeling when a coworker knows how you the expert writes code, and returns a perfect diff with exactly that style. … Bluesky: David Flanagan / @rawkode.dev : Why is Google so bad at product? — This should be in gemini-cli and the remote stuff should be branded Gemini too. — The left hand needs to speak to the right hand. — Do all large companies get this confused?
Context & Ripple Effects
Jules began as an asynchronous agent designed to fix bugs and prepare code changes, then moved into public beta with usage limits in May. This expansion turns that standalone workflow into one developers can invoke from the environments where code is written, tested and discussed.
Google had already brought natural-language coding tasks into terminals with Gemini CLI. Jules Tools extends that developer-surface strategy by making its asynchronous agent available through both a CLI and programmatic integrations.
First-order effects
- Developers and engineering teams can call Jules from terminals, CI/CD pipelines and collaboration tools rather than shifting into a separate product interface.
- The public API lets third-party tools and internal developer platforms trigger and incorporate Jules workflows, broadening the agent's reachable work surface.
Second-order effects
- Coding-agent rivals will face stronger pressure to offer terminal access, APIs and workflow integrations; OpenAI had already positioned Codex as a terminal-based coding agent.
- Teams evaluating Jules can embed agent-driven code work into existing review and delivery processes, making integration fit—not just model capability—a more central procurement criterion.
Third-order effects
- If these integrations become standard, coding agents may compete less as standalone assistants and more as programmable infrastructure inside software-delivery systems.
- That shift would increase the importance of access controls, auditability and human review around agent-triggered changes, particularly when agents operate through shared CI/CD and collaboration channels.
The trend: AI coding tools are evolving from chat-style assistants into embedded, API-addressable agents that operate across the software-development workflow.