TestMu AI, formerly LambdaTest, has introduced Kane CLI, a new terminal-native browser automation tool designed for both human developers and AI coding agents. The San Francisco-based company aims to bridge the gap between AI-driven code generation and the critical step of verifying that code actually works in a browser.
AI agents are rapidly accelerating software development, shipping code and fixing bugs at speeds previously unattainable. However, the development loop has remained incomplete, as no AI agent could independently open a browser and confirm its own output.
Kane CLI is positioned as the verification layer for this new development paradigm. It allows teams to describe browser flows and receive pass/fail results with detailed step traces and screenshots before code submission.
The tool also enables designers and product managers to verify fixes and broken flows without filing tickets, offering shareable evidence links for platforms like Slack or Jira.
For AI agents such as Claude Code, Codex CLI, Cursor, and Gemini CLI, Kane CLI provides the essential validation step, confirming functionality within a local Chrome browser.
Key Capabilities of Kane CLI
- Intent-based browser control: Operates on user intent, bypassing the need for selectors or underlying code.
- Resilient runs: Adapts to changes and completes up to 50 steps per flow, even when encountering initial issues.
- Playwright export: Converts plain English test descriptions into native Playwright code.
- Automated bug discovery: Identifies and flags unexpected behaviors during test execution.
- Vision-based dynamic waiting: Utilizes on-screen visuals to detect loaders and animations, handling complex elements like Canvas and Shadow DOM.
- Assisted OTP and CAPTCHA handling: Pauses for human intervention on sensitive steps like OTPs or CAPTCHAs, allowing the AI workflow to continue.
- Two-way script migration: Converts existing Playwright or Selenium scripts to Kane CLI format and vice-versa.
- Inbuilt Test Manager sync: Automatically saves local test cases remotely with shareable proof.
- CI/CD ready: Supports headless execution in major CI/CD platforms with standard exit codes.
- Contextual authoring: Generates parallel test cases across multiple browser sessions from a single prompt based on provided app context.
Three Ways to Run
Kane CLI offers three distinct modes for both human and AI consumption.
The Interactive TUI provides a full terminal interface for exploration and iteration without arguments.
Headless CLI mode, activated with the --headless flag, is designed for scriptable, display-free runs ideal for CI pipelines.
Agent Mode, enabled by --agent --headless, outputs structured NDJSON that AI agents can natively read to determine subsequent actions.
Asad Khan, CEO & Co-Founder of TestMu AI, stated, "For years, the bottleneck in software was writing the code. Vibe coding removed that. Teams are shipping more software, faster, than at any point in the history of our industry. But it exposed a new bottleneck most teams haven't named yet: trust. Every feature that ships from a prompt is a feature nobody has actually verified. At agentic speed, 'a human will click through it later' is not a plan — it's a liability, compounding at the speed of AI. it's a growing pile of unverified work. That's why we built Kane CLI. One terminal command, a real browser, pass or fail. Software has always trusted the people who wrote it. Now, for the first time, it has to trust the machines. Kane CLI is how trust scales in the agentic era."
TestMu AI, the company formerly known as LambdaTest, is offering bonus credits for the first three months to teams activating a paid plan during the introductory period, providing full access to its cloud capabilities.
Kane CLI is available starting today. Installation is possible via npm or Homebrew.
For those building with AI agents, guidance is available at testmuai.com/kane-cli/agents.md.
Documentation and more information can be found at testmuai.com/support/docs/kane-cli-introduction/ and testmuai.com/kane-cli, respectively.
