██████╗ ██╗ ██╗ █████╗ ██╗ ██╗████████╗██╗ ██╗███╗ ███╗ █████╗ ██╗ ██╗
██╔═══██╗██║ ██║██╔══██╗██║ ██║╚══██╔══╝╚██╗ ██╔╝████╗ ████║██╔══██╗╚██╗██╔╝
██║ ██║██║ ██║███████║██║ ██║ ██║ ╚████╔╝ ██╔████╔██║███████║ ╚███╔╝ ╱|、
██║▄▄ ██║██║ ██║██╔══██║██║ ██║ ██║ ╚██╔╝ ██║╚██╔╝██║██╔══██║ ██╔██╗ (˚ˎ 。7
╚██████╔╝╚██████╔╝██║ ██║███████╗██║ ██║ ██║ ██║ ╚═╝ ██║██║ ██║██╔╝ ██╗ |、˜〵
╚══▀▀═╝ ╚═════╝ ╚═╝ ╚═╝╚══════╝╚═╝ ╚═╝ ╚═╝ ╚═╝ ╚═╝╚═╝ ╚═╝╚═╝ ╚═╝ じしˍ,)ノ
Independent verification for AI-written code.
Discover user journeys, generate and run tests, review changes, and inspect the evidence behind your next release.
Website · Open QualityMax · Documentation · Standalone tools
| Your workflow | Get started |
|---|---|
| Use the platform to manage tests and inspect runs | Web app quickstart |
| Connect your coding agent to hosted QualityMax | MCP quickstart |
| Work locally with standalone tools | 9lives for Playwright execution and page checks, qmax-code, or QA skills |
Hosted workflows use a QualityMax account. Standalone tools have their own prerequisites; model-assisted workflows may require a model provider or coding-agent subscription.
┌─ THE QA LOOP ────────────────────────────────────────────┐
│ │
│ Discover ──> Generate ──> Review ──> Execute │
│ │ │
│ Retain evidence <── Rerun <── Inspect ─┘ │
│ │
└──────────────────────────────────────────────────────────┘
- Turn user journeys into coverage. Generate tests, execute them, and inspect results and available artifacts. Start a first run →
- Verify pull requests. Combine code review, supported native test suites, generated-test feedback, and preview-browser checks. Gate availability and enforcement depend on configuration. PR gates →
- Get specialist review. Use Nexus for attributed findings and shareable review reports. Nexus →
- Evaluate AI applications. Assess conversations and hallucination risks, and explore browser journeys with simulated users. AI validation → · Agentic Eyes →
- Investigate and repair failing tests. Review failure evidence and proposed repairs. Self-healing →
- Bring project context into Slack. Ask project-aware questions and review supported action proposals. Slack → · Memory and grounding →
- Manage and reuse test assets. Organize cases and scripts, work with performance tests, and exchange QTML assets. Test management → · Performance → · QTML →
| Project | Purpose | License / status |
|---|---|---|
| 9lives | Open-source Playwright tools for coding agents: bounded local execution with receipts, static test assessment, page scanning and inspection, locator healing, fix confirmation and fault testing, as a Go CLI, an optional @9l/playwright SDK and an MCP server. 9lives.run |
Apache-2.0; healing, confirmation and fault testing are experimental |
| qmax-mcp | The earlier standalone MCP server for page scanning, inspection and Playwright reproductions; those checks now ship in 9lives (9l scan, 9l inspect, 9l mcp) |
MIT; standalone and optional hosted proxy modes |
| qmax-code | Terminal agent for coding and testing, with standalone and connected workflows | Source available, FSL-1.1-ALv2 |
| Free QA Skills | Reusable QA workflows for supported coding agents | Apache-2.0 |
| 9lives Python | The original Python CLI: local test repair with reruns and reviewable diffs, Cypress and Selenium adapters | MIT; maintenance mode, new Playwright projects use the Go runner |
| Test Grader | Static Playwright test-quality grading | Apache-2.0 |
| Supply Chain Scanner | Python dependency supply-chain checks | Apache-2.0 |
| qmax local agent | CLI and local execution agent for platform workflows | Apache-2.0 |
See each repository for its current requirements and limitations. The hosted platform and individual tools have different licensing and availability.
For integrations that embed QualityMax, see Partner / White-label. For responsive audits and managed flows, see Mobile. Learn how to interpret evidence and receipts.
9lives interactive demo · 9lives first-run example · qmax-mcp demo · 9lives Python repair demo · GitHub Action · n8n workflows · qmax-code releases
For the wider platform, visit the public product guide and documentation. Sovereign deployment is available as a private preview.
To contribute to a tool, start with its repository's contribution guidance and issue tracker.