TestDriver
By TestDriver. Category: Agentic end-to-end testing. Homepage
TestDriver's documentation describes it as a testing tool in which an AI agent writes and runs tests against web and desktop applications in TestDriver sandboxes. Tests are created from natural-language prompts or code and are stored as Vitest files that depend on the testdriverai SDK. The docs describe auto-healing, and a GitHub App responds to @testdriverai mentions. CI guides cover GitHub Actions, GitLab, CircleCI, Azure DevOps and Jenkins. Its init command installs an MCP server, agent and skills for Claude Code, Cursor, VS Code, Windsurf, Codex and Zed. A self-hosted option is documented for enterprise customers.
Description written from the vendor's documentation.
Status
Status: Active
Facts
| Fact | Value | Evidence | Source | Verified on |
|---|---|---|---|---|
| Status | Active | Documented | testdriver.ai | |
| AI capabilities | Documented | docs.testdriver.ai | ||
| Test creation | Documented | docs.testdriver.ai | ||
| Test output | Proprietary runner | Documented | docs.testdriver.ai | |
| Surfaces | Documented | docs.testdriver.ai | ||
| Hosting | Documented | docs.testdriver.ai | ||
| Pricing model | Documented | testdriver.ai | ||
| Free tier | Not documented | Not documented | None found | |
| CI integrations | Documented | docs.testdriver.ai | ||
| Self-healing | Documented | Documented | docs.testdriver.ai | |
| Agent integration | TestDriver MCP server, agent and skills installed by `npx testdriverai init` for Claude Code, Cursor, VS Code (Copilot), Windsurf, Codex and Zed; GitHub App responding to @testdriverai mentions | Documented | docs.testdriver.ai |
Published prices
| Plan | Price | Billing |
|---|---|---|
| Pro | $20/ month / seat | monthly per seat, plus $0.14 Per Testing Minute |
Vendor-stated, verbatim
What TestDriver says about its product, quoted as published and not confirmed in documentation.
“The cache makes the test run a maximum of **1.7x faster**.”
Alternatives
Tools in the same category (Agentic end-to-end testing), ordered by the number of shared creation and surface values, then by name. Each line lists documented facts that differ from TestDriver.
- QA Wolf: documents output Mixed
- Autify: documents output Mixed; creation by recording; a free tier
- Heal: documents output Mixed; a free tier
- Katalon: documents creation by recording; a free tier
- Robonito: documents creation by autonomous from URL, recording; a free tier
- UiPath Test Cloud: Agentic end-to-end testing; no documented difference in output, creation, hosting or free tier
Questions
Is this page a review of TestDriver?
No. This page is a fact sheet. Each fact shows its evidence level, its source and the date it was checked. We do not rate or rank tools.
Does TestDriver have a free tier?
A free tier is not documented for TestDriver as of 2026-09-24.
What test output does TestDriver produce?
Output: Proprietary runner. This is documented, checked on 2026-09-24.
Is TestDriver still available?
Status: Active, checked on 2026-09-24.
Sources
- TestDriver homepage Opened
- TestDriver pricing Opened
- TestDriver docs home Opened
- TestDriver docs index (llms.txt) Opened
- What is TestDriver? Opened
- Integrating into CI/CD Opened
- Self-Hosted (Enterprise) Opened
- Self-Healing (performing actions) Opened
- Prevent: auto-healing and GitHub integration Opened
- Self-Learning (caching) Opened
- Configure Your Agent (CLI quickstart) Opened
- Spawning a sandbox (provision) Opened
- testdriverai/testdriverai GitHub repository Opened
Spotted an outdated or wrong fact about TestDriver? Send the source and we apply it at the next monthly update.