cursor · Nov 9, 2026 · Nov 9, 2026
@@ -1,6 +1,17 @@
 ---
 name: e2e
 description: Writes, runs, and debugs end-to-end tests for web and mobile apps with e2e, the agentic test runner, which mixes agent.act goals with exact locator assertions on a Playwright browser or an iOS or Android device. Also drives a project's running app live over the e2e MCP server to look at a screen, check a UI change, take a screenshot or record a video, and try a locator; and runs bug bashes, parallel e2e explore runs with each bug proven by a repro test. Use when a project depends on e2e or has an e2e.config.ts, when asked for end-to-end, browser, mobile, or agentic UI tests, to check a change in the running app, to QA or bug bash an app or branch, or when an e2e run fails.
+metadata:
+  openclaw:
+    homepage: https://e2e.com
+    requires:
+      bins:
+        - node
+        - npx
+    envVars:
+      - name: AI_GATEWAY_API_KEY
+        required: false
+        description: Model key for agent steps when the project's e2e config uses the Vercel AI Gateway; other providers read their own key.
 ---
 
 # e2e: agentic end-to-end tests
@@ -108,8 +119,10 @@ test('a member upgrades to Pro', async ({ app, agent, screen, browser }) => {
    Exact values go through `screen`: a sign-in form in a setup test, a field
    that must receive one specific string, a count that must be one number.
 4. Run only the test you are working on:
-   `npx e2e run tests/<feature>.e2e.ts:<line>`, the line its `test(` opens
-   on, with the flags in "Rerunning a test while you fix it" below. Agent
+   `npx e2e run tests/<feature>.e2e.ts:<line>` (over MCP,
+   `run_tests {files: ["tests/<feature>.e2e.ts:<line>"]}`, the flags as
+   camelCase fields such as `retries: 0`), the line its `test(` opens on,
+   with the flags in "Rerunning a test while you fix it" below. Agent
    steps need a model in the config and that provider's authentication (a
    saved subscription login, an API key); a local endpoint may need none.
    Tests without agent steps need no model.
@@ -195,7 +208,8 @@ lines you changed:
 - In a project without e2e, set it up when the user requests it. Otherwise
   ask before adding a test framework, and name any existing framework.
 - Live MCP tools need the project's own `e2e mcp` server registered in the
-  client (topic `mcp`); writing and running tests use the CLI directly.
+  client (topic `mcp`). Tests run with the CLI, or with the server's
+  `run_tests`, the same run without a shell.
   Keep a session open across runs while you write or fix a web test; topic
   `mcp` says when a run can go beside it, and what to do on
   `APP_ALREADY_RUNNING`.

@@ -49,14 +49,14 @@ under `executor`. `openai.decisionModel('gpt-6-luna')` from `@ai-sdk/openai`
 4.0.90 or later works too and takes the `vision: true` screenshot.
 Tests stay plain language with no params. The text model writes field values
 and upload paths. See the shipped `docs/decision-models.mdx` or
-[the online guide](https://e2e.tester.army/docs/decision-models) for setup,
+[the online guide](https://e2e.com/docs/decision-models) for setup,
 gates, and limits.
 
-A coding agent the user already signs in to runs agent steps over ACP:
+A coding agent the user already signs in to runs agent steps over the Agent Client Protocol:
 install `@e2e-dev/acp` and its adapter, and put
 `claudeCode({ model: 'sonnet' })` or `codex()` under
-`executor`; no model provider is needed. See the shipped `docs/acp.mdx` or
-[the online guide](https://e2e.tester.army/docs/acp).
+`executor`; no model provider is needed. See the shipped `docs/subscriptions.mdx` or
+[the online guide](https://e2e.com/docs/subscriptions#use-your-coding-agent).
 
 - Pass a model instance, not a string (`INVALID_CONFIG`).
 - An agents entry is one plain object of `model`, `judge`, `system`,
@@ -154,6 +154,9 @@ The tools, one per engine action. Name the target as the screen names it
   native file chooser; outside it or hidden (`.env`) is `POLICY_DENIED`.
 - `navigate`: a URL or app-relative path.
 - `back`: browser history or in-app back.
+- `resize_viewport`: set the viewport size (200 to 4096 CSS px) to check a
+  layout at phone, tablet, or desktop width. Engines with `session.resize`,
+  such as web.
 - `type_secret`: a declared secret by name; plaintext never reaches the model.
   Offered only when the step declares secrets and the engine can fill them.
 - `screenshot`: attach viewport pixels; every later result then carries one.
@@ -241,6 +244,10 @@ For a debug view that pairs each action with the screen it started on, give
 each capture to `onAction` by `index`, which restarts at 1 in each `act` call.
 It runs before replayed actions too, does not wait for the screen to settle,
 and a failed action gets a capture but no `onAction`.
+Narrow `onAction`'s event on `action` for coordinates, replay included:
+`point` for `tapAt`/`hoverAt`, `from`/`to` for `dragAt`. They are in the
+same space as `screen.tapAt` and `boundingBox()`, rounded and clamped to the
+viewport. Mobile engines have no `hoverAt` or `dragAt`.
 
 `vision` on a judgment picks the evidence: `false` (default) the tree;
 `true` the tree plus a masked screenshot; `'only'` the screenshot alone, for
@@ -428,7 +435,7 @@ make no model calls.
   `executor`; the runner still owns observations, actions, budgets, and the
   report, and `system` or `tools` beside `executor` is `INVALID_CONFIG`.
 
-Reference: https://e2e.tester.army/docs/agents
+Reference: https://e2e.com/docs/agents
 
 ## In CI
 

@@ -29,7 +29,7 @@ default.
      timeout, changing values at a race;
    - what the app logged meanwhile (`✗ network error: POST /api/save 500`,
      `✗ uncaught error: ...`, `⚠ console warning: ...`, `ℹ console: ...`);
-     `❯ app 2 errors` in the terminal says there are some, and the page's
+     `❯ app 2 errors` in the terminal quotes the first 3 errors, and the page's
      `## App log` lists up to 50 with their step, errors and warnings
      first, counting the rest;
    - where the page went (`↪ navigated to /login`, a new tab the app

@@ -72,8 +72,8 @@ configured secrets are redacted (topic `writing-tests`).
 | `--target <id>` | first configured target | The one target to explore. |
 | `--agent <name>` | `default` | Build the explorer from another configured agent (`agents.<name>`). |
 | `--session <name>` | none | Run the setup that saves this session, then explore with it restored. |
-| `--max-steps <n>` | 8 (1 to 12) | Exploration steps at most. |
-| `--timeout <ms>` | 600000 (180000 to 900000) | Wall clock; the last minute is for the assessment. |
+| `--max-steps <n>` | 8 (1 to 100) | Exploration steps at most. |
+| `--timeout <ms>` | 600000 (180000 to 7200000) | Wall clock; the last minute is for the assessment. |
 | `--headed`, `--reporter`, `--output`, `--debug`, `--ai-trace`, `--trace [mode]`, `--video [mode]` | as `run` | Same meaning as for `e2e run`. The exploration is one attempt, so a retry mode (`on-first-retry`, `on-all-retries`) records nothing (the run's notice says to pass `--trace on`); put the goal before a bare `--trace` or `--video`. |
 
 Per-step action and model-call budgets default to 40 each;