Background agents can write code. But you still need to verify the result. Lark preview deployments let you open a real app environment for any PR, test the change directly, and check side effects before merging. Docs: docs.getlark.ai/preview-depl…

Lark is sponsoring AI Dev Summit this week. If you’re here, come say hi. We’d love to chat about faster releases, fewer regressions, and AI-powered testing.

One of our customers went from zero e2e test coverage to 100 tests defined in Lark in under an hour. Less than a week later, Lark has run close to 5,000 tests across their critical flows — surfacing API flakiness and helping them catch regressions before users do. Try the Lark MCP or book a demo: getlark.ai

E2E tests shouldn’t break every time your UI evolves. Today, one of Lark’s own E2E tests failed after we shipped a UI change. Lark summarized the failure, identified it as a test issue rather than an app bug, and repaired the test automatically. The test has been passing consistently since.

Over 100 QA reports generated since launch. We’re seeing real issues uncovered across SDKs, APIs, and dashboards—helping teams catch and fix problems before users report them. Fast, structured, reproducible.

Create Linear issues automatically when tests fail or bugs are discovered. Catch bugs → create issues → fix faster. No missed failures. No manual triage.

We just shipped QA Reports. Deploy AI agents to test your product, uncover issues, and get back a structured report with reproducible test cases. We ran it on the Vercel Sandboxes API. It mapped test areas, executed flows, and surfaced real issues in minutes. Try it: getlark.ai Example report (Vercel Sandboxes API): tests.getlark.ai/a/kiwi/qa-r…

We just launched Repairs at Lark. Ship a UI change. End-to-end tests break. Lark's QA agents automatically fix them. No more test maintenance bottlenecks. Just turn on auto-repairs and keep shipping. Try it → getlark.ai

Run agents like Claude Code in any sandbox. We just open sourced runtimeuse.com - a framework for running agents in sandboxes without building all the infra yourself. Agents sandboxes are powerful. The hard part is everything around them: - streaming messages - file exchanges - cancelling jobs mid-run, etc We hit this while building @getlark, so we open sourced our runtime. Try it: `npx -y runtimeuse --agent=claude` Feedback welcome 👇

Writing code is now cheap. Review and testing are the bottleneck. We solve that.

Shipping fast shouldn’t mean breaking production. With Lark, you can write and run an end-to-end test in under 2 minutes. Run it in staging or production. Catch breaking changes before users do. Try it: getlark.ai