TestMu AI (Formerly LambdaTest) is a Full-Stack Agentic AI Quality Engineering platform that empowers teams to test intelligently and ship faster. Engineered for scale, it offers end-to-end AI agents to plan, author, execute, and analyze software quality. AI-native by design, the platform enables testing of web, mobile, and enterprise applications at any scale across real devices, real browsers, and custom real-world environments.

TestMu AI is used by 3M+ users and 18,000+ enterprises in 132 countries, and was named a Leader in the 2026 Gartner Magic Quadrant for Agentic Software Quality Assurance Platforms.

Product	What it tests	Install	Docs
Kane CLI	Web, API and mobile apps, from a terminal or a coding agent	`npm install -g @testmuai/kane-cli`	Docs
Agent Testing	Chat, voice, phone, IVR, video and image agents	`pip install agent-testing-cli`	Docs
Agent Assurance	Coding, workflow, back-office and multi-agent systems that act on tools, files and APIs	`brew install lambdatest/rook/rook`	Docs

Kane CLI

Kane CLI is an AI-powered command-line testing agent from TestMu AI that runs browser and native mobile app tests written in plain English, from your terminal, IDE or CI pipeline, and returns an evidence-backed pass or fail. It works with Claude Code, Cursor, Codex CLI, Gemini CLI and GitHub Copilot.

Tests five surfaces: web in real Chrome, API, Android Emulator, iOS Simulator and real devices
Needs no selectors, scripts or test framework, and any recorded flow exports to Playwright code
Saves every run's steps, screenshots, console and network logs and verdict as an evidence file
--agent mode prints structured NDJSON for coding agents; exit codes are 0 pass, 1 fail, 2 setup or auth error, 3 timeout
Runs headless in GitHub Actions, GitLab CI, Jenkins and Bitbucket Pipelines
Free to install; runs use KaneAI credits, and new users get 10,000 free credits for 30 days

# npm (Node.js 18+)
npm install -g @testmuai/kane-cli

# Homebrew (macOS and Linux, no Node.js needed, installs Chrome)
brew install LambdaTest/kane/kane-cli

# Shell installer (any POSIX system)
curl -fsSL https://raw.githubusercontent.com/LambdaTest/kane-cli/main/install.sh | sh

# Skill for Claude Code, Codex CLI or Gemini CLI
npx @testmuai/kane-cli-skill

# First run
kane-cli login
kane-cli run --url https://example.com "Click the 'More information' link and verify the page loads"

For Cursor, Copilot and other coding agents, point the agent to AGENTS.md.

Coding agent guides | GitHub | Launch blog

Agent Testing

TestMu AI Agent Testing is an AI agent testing platform for chat, voice, phone, IVR, video and image agents. Autonomous AI evaluators hold real conversations, phone calls and video calls with your agent, score every response for hallucination, bias, toxicity and compliance gaps, and return a Green, Yellow or Red go-live verdict before and after release.

Runs 15+ specialized AI testing agents in parallel
Scores chat and voice agents on 9 quality metrics and phone agents on 30+ call metrics, each with a confidence level
Generates 60 to 100+ scenarios from PRDs, docs, or connected Jira and Confluence, alongside a library of hundreds of prebuilt scenarios
Connects a chat agent with no SDK: upload the docs, define intended behavior, and point it at the endpoint or phone number
Tests inbound and outbound phone agents with voice selection and 15 background noise presets, and batch-analyzes production call recordings
Scores AI-generated images against prompts, brand rules and technical specs, up to 50 images per batch

# Agent Testing CLI (Python 3.10+)
pip install agent-testing-cli
agent-testing-cli login
agent-testing-cli projects

For coding agents, point the agent to agents.md.

Video agent testing

Agent Assurance

Agent Assurance is an end-to-end testing product from TestMu AI for AI agents that take actions. Its Rook CLI derives a test suite from your codebase, PRD or policy docs, runs your agent against it for real, grades each criterion on the tool calls, file changes and API effects the run produced, and reports anything it could not verify separately.

Covers coding, workflow, back-office and multi-agent systems, including n8n agents over a webhook
One generated suite covers functional, regression, smoke, automation, continuous and adversarial agent testing
Generates adversarial scenarios for prompt injection, instruction override and tool misuse by default
Returns three verdicts, pass, fail and unable to verify, and keeps unverifiable criteria out of the pass rate
Runs headless in CI with a JSON report and published gate recipes for GitHub Actions, Jenkins and Argo CD
Keeps scenarios and results in your repository as plain files; 14-day free trial

# Homebrew
brew install lambdatest/rook/rook

# npm (Node.js 20+)
npm install -g @testmuai/rook@latest

# Shell installer
curl -fsSL https://raw.githubusercontent.com/LambdaTest/rook/main/install.sh | bash

# Windows PowerShell
npm.cmd install -g @testmuai/rook

# Skill for Claude Code, Codex CLI, Gemini CLI and more (install alongside the CLI)
npx @testmuai/rook-skill

Rook CLI | Launch blog | Multi-agent testing guide | Pricing

FAQ

What is TestMu AI?

TestMu AI (Formerly LambdaTest) is a Full-Stack Agentic AI Quality Engineering platform that empowers teams to test intelligently and ship faster. It offers end-to-end AI agents to plan, author, execute, and analyze software quality, and tests web, mobile, and enterprise applications across real devices, real browsers, and custom real-world environments. Its products include KaneAI, Kane CLI, Agent Testing, Agent Assurance, HyperExecute, Real Device Cloud, SmartUI and Test Manager.

Is TestMu AI the same company as LambdaTest?

Yes. LambdaTest rebranded as TestMu AI on January 12, 2026, moving from cross-browser testing to a full-stack agentic AI quality engineering cloud.

What is the difference between Agent Testing and Agent Assurance?

Agent Testing evaluates agents that talk to people (chat, voice, phone, IVR, video and image agents) by holding conversations with them and scoring each response. Agent Assurance tests agents that act (coding, workflow, back-office and multi-agent systems) by checking the tool calls, files and API changes each run produced.

How is Kane CLI different from Agent Assurance?

Kane CLI verifies the software a developer or coding agent builds by running user flows in a real browser or mobile app. Agent Assurance verifies the behavior of the AI agent itself.

How do I use Kane CLI with Claude Code, Cursor or Codex?

Install Kane CLI, then add the skill with npx @testmuai/kane-cli-skill for Claude Code, Codex CLI or Gemini CLI, or point Cursor and Copilot to AGENTS.md. Run with --agent --headless and gate on the exit code.

Can these tools run in CI/CD?

Yes. Kane CLI runs headless with standard exit codes, the Agent Testing CLI supports --json and --yes for pipelines, and Agent Assurance publishes gate recipes for GitHub Actions, Jenkins and Argo CD.

Is there a free trial?

Kane CLI is free to install and includes 10,000 free credits for 30 days. Agent Assurance has a 14-day free trial. Agent Testing can be started free at testmuai.com/register.

Links