Assert
@Pixel-Funnel
Run real Playwright E2E tests from your AI coding agent — no Playwright knowledge required. Describe what to test in plain English, Assert writes and executes the test against your live app, and reports back step-by-step results.
Overview
What is Assert?
Assert MCP connects your AI agent (Claude, Cursor, Windsurf) to the Assert testing platform, enabling test scenario generation from plain English, execution against live apps via Playwright, and pass/fail results with step-level details and screenshot URLs — all without writing test code.
How to use Assert?
Install the MCP server and configure it to use the provided tools: assert_list, assert_generate, assert_run, and assert_status. A free tier is available at assert.click. No Playwright knowledge is required.
Key features of Assert
- Connect AI agents (Claude, Cursor, Windsurf) to the Assert testing platform.
- Generate test scenarios from plain English descriptions.
- Run tests against live applications using Playwright.
- Receive pass/fail results with step-level details and screenshot URLs.
- No test code required.
- Free tier available at assert.click.
Use cases of Assert
- QA teams generate test scenarios from plain English without coding.
- Developers integrate AI agents to run tests against live apps.
- Quickly obtain pass/fail results with screenshots for debugging.
- Ad-hoc testing by describing a test in markdown.
FAQ from Assert
What tools does Assert MCP provide?
It provides assert_list to list saved scenarios, assert_generate to create a scenario from a description, assert_run to execute a scenario or ad-hoc test, and assert_status to poll for step-level results.
Do I need Playwright knowledge to use Assert?
No. Playwright runs in the background; you only describe tests in plain English or markdown.
Which AI agents are supported?
Assert MCP connects to Claude, Cursor, and Windsurf.
Is there a free tier for Assert?
Yes, a free tier is available at https://assert.click.
How do I get test results?
Use the assert_status tool to poll for step-level pass/fail results, including screenshot URLs.
