Nov 18, 2026
|As part of a new direction for Stitch, we’re introducing a new experiment, Stitch Loop: a long-running product agent that connects to your codebase and live metrics to continuously improve what you’ve shipped.
Justin Toupin
Director, Product Management, Google Labs
Share
Listen to article
[[duration]] minutes
This content is generated by Google AI. Generative AI is experimental
We are introducing Stitch Loop, a product agent that connects to your codebase and live metrics to continuously optimize your software. It identifies issues, suggests fixes, and opens pull requests to help your team maintain product quality around the clock. You can join the waitlist today at stitch.google.com/loop/waitlist to get started with these automated tools for security, latency, and accessibility.
Summaries were generated by Google AI. Generative AI is experimental.- Justin Toupin introduces Stitch Loop, a product agent that monitors your codebase and metrics continuously.
- This tool identifies areas for improvement, writes code in a sandbox, and submits pull requests.
- You maintain full control over all changes because you must approve every pull request manually.
- The system includes dedicated loops for cybersecurity, latency, and accessibility to help your lean team.
- You can join the global waitlist today to receive priority access for your development projects.
We are introducing Stitch Loop, a new tool from Google Labs that helps you maintain and improve your software after it launches. It connects to your code and live data to find bugs, fix security issues, and boost performance around the clock. You simply review the suggested changes before they go live. This helps your team focus on building new features while the agent handles the ongoing work of keeping your product running well.
Summaries were generated by Google AI. Generative AI is experimental.Explore other styles:
Coding agents let you build and ship new features fast, but their job ends the moment your pull request merges. Once code hits production, the real work starts: patching vulnerabilities, speeding up checkout, and diagnosing signup drop-offs buried in noisy feedback. For lean teams, balancing post-launch maintenance with continuous improvement is a constant challenge, yet it’s also what drives some of the world’s most successful products.
Today, we are introducing Stitch Loop, a long-running product agent from Google Labs that connects to your codebase and live metrics, finds what needs fixing, and keeps working 24/7 on your product until your goals are met.
Millions of designers, developers, and founders already use Stitch to turn ideas into working interfaces in minutes. Together, Stitch and Stitch Loop are steps toward our broader ambition for Stitch: creating agents to help lean teams tackle the hardest parts of creating great products from ideation through continuous optimization.
See how it works
You start by connecting GitHub and your product context (logs, user behavior, feedback, etc) with over 180 integrations, custom MCP servers, or local CLI uploads. Grounded in that context, Stitch Loop runs evergreen loops to continuously optimize toward the measurable outcomes you care about. You get three default loops out of the box and can add your own anytime. Each loop runs through four steps:
- Suggestions: Stitch Loop surfaces opportunities to make progress towards your goals, grounded in evidence from your product.
- Solutions: Every suggestion comes with one or more solutions— each with tradeoffs and scoping, along with visual design options where applicable.
- Build: Once you choose a direction, Stitch Loop writes the code in an isolated cloud sandbox and opens a GitHub pull request for your approval. When relevant, the PR will contain screenshots and videos of the work Loop did.
- Reflect: Stitch Loop monitors live product data as it works toward your goal, learning from every positive or negative result to sharpen the next round of insights. Once the goal is met, it keeps watching in case the metric slips.
You always have the final call. Nothing merges until you review the PR.
Automate codebase security
Stitch Loop comes with built-in loops for cybersecurity, latency, and accessibility, plus the ability to create custom loops like improving checkout conversion.
Security is often the hardest gap to cover for lean teams shipping quickly. Stitch Loop includes a dedicated security loop that pairs Gemini 4 Carbon's cyber security capabilities with CodeMender, Google DeepMind's autonomous agent that improves code security. Together, they scan your codebase for security vulnerabilities before they become incidents. Stitch Loop flags each issue it finds with evidence and opens a PR with a proposed fix for your review.
Use it with your agents and existing workflows
You can use Stitch Loop as a service layer for coding agents, like Google Antigravity, or bring it into the places where your team already works, like Slack.
Whatever coding agent you prefer can pull persistent product context from Stitch Loop via the Stitch CLI and MCP to power mixed-model setups and fully autonomous improvements in your existing workflow.
In Slack, just use @stitch to get daily insight digests and create new loops.
Join the waitlist
The Stitch Loop waitlist opens today for all users, globally, with eligible Google AI Ultra users getting priority access. Join at stitch.google.com/loop/waitlist and share your feedback in the Google Labs Discord.
Get the latest news from Google in your inbox
Sign up for our newsletters with product updates, event information, special offers, and more.
Done. Just one step more.
Check your inbox to confirm your subscription.
You can also subscribe with a .
Your information will be used in accordance with Google's privacy policy. You may opt out at any time.
Posted in:
AI 热点