AI-Assisted & Agentic Testing
LLM-driven tools that let agents drive browsers, generate tests from traffic or run natural-language test suites.
7 tools compared
Download trends
Weekly downloads for the 3 most-downloaded tools in this category. Hover for exact values, or switch to log scale to compare tools of very different sizes.
- Playwright MCP
- Stagehand
- Midscene.js
Comparison
Click a column to sort. “#” is the curated ranking, with the most widely used first.
| 4 |  Browser UseTrending Python library that lets LLM agents see and operate websites in a real browser. | 121k | 1.9M | -14% | 1d ago | 84 |
| 4 |  Midscene.jsRising Vision-driven UI automation that lets you write web, Android and iOS tests as natural-language steps. | 15.5k | 71.1k | +47% | 1w ago | 67 |
| 5 |  Stagehand Browserbase's TypeScript framework mixing code and natural-language actions for reliable browser agents. | 26k | 2.9M | +64% | 1d ago | 80 |
| 5 |  KeployRising Records real API traffic via eBPF and turns it into regression tests and mocks automatically. | 19.5k | n/a | n/a | 4d ago | 78 |
| 6 | QA Wolf Managed service that builds and maintains Playwright E2E suites with AI plus human QA engineers. | No public repository or package of its own to measure | ||||
| 7 | MomenticNew AI testing platform where E2E tests are written in plain English and self-heal as the UI changes. | No public repository or package of its own to measure | ||||
| 8 |  Playwright MCP Microsoft's MCP server that lets agents drive a real browser via Playwright accessibility snapshots. | 39.5k | 10.5M | +28% | today | 85 |