TAMASH · Forge, Shield and RunSightsThree products, one portal, your existing Playwright or Cucumber tests. Plan and sign off releases in Forge, keep your automation running with Shield, and read every run in RunSights.
For developers, QA teams, and AI coding agents.
One portal, one project key, the same people. Shield includes RunSights; Forge includes both.
Test cases, cycles, executions and reports, connected to the repository. The agent explores the application and drafts tests while the team plans and signs off releases.
A renamed button no longer fails the nightly run. Shield heals the locator while the test runs and turns the fixes worth keeping into code changes.
locator('#pay-btn')getByRole('button', { name: 'Pay now' })HealedEvery run from laptops and CI in one place: what newly failed, what got fixed, what is flaky, and why.
An npm token and a project key, and your tests run as before.
$ npm install --save-dev @vibetestq/tamash-shield @vibetestq/tamash-runsights
$ npx tamash-shield init
[changed] package.json: installed @vibetestq/tamash-runsights
[changed] playwright.config.ts: added the reporter
[changed] .env: TAMASH_LICENSE_KEY, paste your project key
$ npx tamash-shield login
Signing in dana-macbook. Open this address and approve the machine when the code reads 7KQ2-M9FD:
https://tamash.vibetestq.com/connect?code=7KQ2-M9FD
Signed in as dana@example.com: dana-macbook now has its own access token.$ npx playwright test
✓ checkout.spec.ts › the cart lists the items
✓ checkout.spec.ts › a coupon code can be applied
[self-healer] checkout.spec.ts:18 — locator.click "Pay now" -> HEALED [by TAMASH, suggested="getByRole('button', { name: 'Pay now' })"]
✓ checkout.spec.ts › the payment step opens
✓ account.spec.ts › a saved address is shown
8 passed (6.2s)
TAMASH RunSights: run sent (8 tests)The full guide covers CI and Cucumber.
Test definitions and reusable automation stay in the repository, available to the people and agents doing the work.
Quality assets are reviewed alongside software changes, with version history and a shared project context.
Execution results and maintenance history decide what to fix, verify or run next.
An agent explores the application in Forge, drafts executable specifications and proposals, generates reusable tests, and runs them with Shield. The results are in RunSights before the pull request is opened.
Open the run in RunSights, read the likely cause of a failure, fix it, and run again. A locator that changed on the way is healed by Shield rather than failing the build.
Plan cycles in Forge from what changed, assign tests, review healing and maintenance, and sign the release off against accepted test versions with the evidence attached.