Skip to content
MCP server

TestWatch

Run end-to-end tests, read why they failed and generate Playwright tests from an agent

Install
Pricing
Freemium
Platforms
Web, API
Runs
Remotely
Version
1.1.0
Listing updated
2 days ago

About TestWatch

TestWatch is a freemium MCP server for developer tools. It runs as a hosted remote server, so there is nothing to install locally.

TestWatch gives a coding agent a feedback loop on the running app. In the docs' example, an agent checks the checkout on a preview URL: it reads status, finds tests tagged checkout, runs them against the pull request's preview, waits for the verdict and reads why a failure happened, with category, failing step, assertion diff, logs and screenshot. It then fixes the code and re-runs only the failed tests.

Permissions are granular. An owner or admin must switch on AI agent access, which is off by default, and on the approval page you choose read results (always), start and cancel runs, and generate tests, optionally limited to one app. Deleting tests, apps or environments is not possible over MCP. New tests are written as Playwright tests from a plain-English goal, take 3 to 15 minutes, and need your confirmation because they use a generation.

The docs warn that error text and page content come from the site under test and may carry injected instructions, so such text is wrapped in marked untrusted-data blocks, which lowers but does not remove the risk. Limits are 10 runs a minute and 10 generations an hour per connection. MCP access starts on the Starter plan.

Install

Run in a terminal

claude mcp add --transport http testwatch https://mcp.testwatch.dev/mcp

Add to Settings → Connectors → Add custom connector

https://mcp.testwatch.dev/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "testwatch": {
      "url": "https://mcp.testwatch.dev/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "testwatch": {
      "type": "http",
      "url": "https://mcp.testwatch.dev/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "testwatch": {
      "serverUrl": "https://mcp.testwatch.dev/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "testwatch": {
      "type": "http",
      "url": "https://mcp.testwatch.dev/mcp"
    }
  }
}

Tools this server provides (9)

Taken from the project documentation; the running server may differ.

testwatch_status
Apps, latest results per environment, review backlog and plan usage
testwatch_find_tests
Search an app's tests by text, tag or status
testwatch_get_test
A test's goal, steps, status, recent results and optionally its code
testwatch_run_tests
Start a run and wait up to 45 seconds
testwatch_get_run
Wait for progress, then the verdict and failures
testwatch_get_result
Why one test failed
testwatch_cancel_run
Cancel tests of a run that have not started
testwatch_review_inbox
Real bugs, suggested fixes and blocked tests waiting for a person
testwatch_generate_test
Write a new Playwright test from a plain-English goal

What it does

Run and wait
Start runs by tag, id or preview URL and wait for the verdict.
Failure diagnosis
Category, failing step, assertion diff, logs and optional screenshot.
Test generation
Write a Playwright test from a plain-English goal, with confirmation first.
Review inbox
Real bugs, suggested fixes and blocked tests awaiting a person.
Scoped permissions
Read, run and generate permissions chosen at approval, per app.
Untrusted-data wrapping
Site-originated text is marked as data for the agent.

Good for

  • Check a pull request's preview deployment before merging
  • Ask an agent why last night's run failed
  • Generate an end-to-end test for a new feature
  • Re-run only the failed tests after a fix
  • Triage the review inbox from your editor

Pricing

Freemium, from $79 a month

Free for 5 tests; Starter $99/mo ($79 billed yearly); Team $299/mo; Enterprise from $799/mo; MCP needs Starter

Prices change. Check testwatch.dev before you buy.

TestWatch FAQ

Which plans include MCP access?

Starter and up, per the docs; Starter is $99 a month, or $79 billed yearly.

Is agent access on by default?

No. An owner or admin turns it on under Settings, AI coding agents.

Can an agent delete tests?

No. Deleting tests, apps or environments stays in the dashboard.

Can I use an API key instead of sign-in?

Yes, create one under Settings, API keys and send it as a Bearer token.

Are runs billed?

Runs count against monthly run allowances, and going over is never billed.

Is TestWatch free?

TestWatch has a free plan, with paid plans for more usage or features. Paid plans start at $79 a month.

How do I add TestWatch to Claude Code?

Run this command in your terminal: claude mcp add --transport http testwatch https://mcp.testwatch.dev/mcp

What are the best alternatives to TestWatch?

The closest alternatives to TestWatch are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to TestWatch

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools