io.github.trajectiq-ai/E2E

What it mainly works on, as our AI model read its tool descriptions: files and code on the machine it runs on; a web browser or desktop applications.

5 of 11 tools are described as changing something (creating, editing, deleting, sending or running). For 1 the reading was uncertain, so no mark is shown.

• run-teston 1 other serverruns code or commandsdescribed as changing data

  “Run the project's Playwright tests: stats plus each failure's file:line, kind and message.”

  Publisher declares: read-only no · destructive yes · idempotent yes · open-world yes

  14 inputs

  • projectRootstring“Project directory; defaults to the server working directory”
  • testFilesarray“Test files/directories relative to the project root; "file:line" is supported.”
  • grepstring“Only run tests whose title matches this regex”
  • browserstring“Playwright project name to run (matched against playwright.config projects)”
  • headedboolean“Run with a visible browser window”
  • timeoutMsinteger“Hard wall-clock limit for the whole run; the process tree is killed past it (default 120000)”
  • testTimeoutMsinteger“Per-test timeout passed to Playwright”
  • workersinteger
  • retriesinteger
  • retryOnFailureboolean“Auto retry failing tests once before reporting failure (default true; ignored when retries is set)”
  • lastFailedboolean“Only re-run tests that failed in the previous run (Playwright --last-failed) — the fast fix → re-run loop”
  • configstring“playwright.config path or 1-based index when the project has several configs”
  • argsarray“Extra Playwright flags from an allowlist, as --flag or --flag=value (e.g.”
  • backgroundboolean“Return a run id at once; poll get-run-status”
• get-run-statuson 1 other serverreads or searches

  “Progress (tests done/total, failures so far) or the final result of a run-test background run; the result becomes the last run for get-failure.”

  Publisher declares: read-only no · destructive no · idempotent yes · open-world no

  3 inputs

  • runIdstring
  • waitSecondsinteger“Wait up to this long for it to finish”
  • cancelboolean
• get-failureon 1 other serverreads or searches

  “Analyze one failure from the most recent run-test execution: full message and code frame, expected vs actual, stack trace, failed network requests (4xx/5xx and dead endpoints), console errors, failur…”

  Publisher declares: read-only yes · idempotent yes · open-world no

  2 inputs

  • indexinteger“1-based failure index from the last run (default 1)”
  • projectRootstring“Only used to re-read the stored report”
• suggest-fixon 1 other serverruns code or commandsdescribed as changing data

  “Patch a failure from the last run: a drifted locator is matched against the DOM at failure (trace, or live url) and replaced with one proven unique; changed copy gets the new expected text.”

  Publisher declares: read-only no · destructive yes · idempotent no · open-world yes

  10 inputs

  • indexinteger“1-based failure index (default 1)”
  • projectRootstring
  • urlstring“Heal against this live page instead of the trace snapshot”
  • storageStatestring“Logged-in session: storageState JSON inside the project, e.g.”
  • headersobject“Extra HTTP headers, e.g.”
  • applyboolean“Write the fix (default: diff only)”
  • verifyboolean“Re-run after applying; revert if still failing (default true)”
  • allowExpectationUpdateboolean“Allow applying changed expected text (can hide regressions)”
  • configstring“playwright.config path or 1-based index”
  • timeoutMsinteger“Verification run limit (default 120000)”
• inspect-pageon 1 other serverreads or searches

  “Open a URL headlessly and return its elements with Playwright locators proven unique on the page (role + name first), CSS selectors, visibility, text and console messages.”

  Publisher declares: read-only yes · idempotent yes · open-world yes

  13 inputs, 1 required

  • urlstringrequired“Full URL of the page to open (http/https)”
  • projectRootstring“Project whose Playwright launches the browser (default: server root)”
  • selectorstring“CSS selector: inspect matches instead of the whole DOM”
  • viewstring“'locators': compact map of interactive elements (cheapest); 'elements' (default): DOM inventory”
  • waitForstring“Wait for this selector first”
  • storageStatestring“Logged-in session: storageState JSON inside the project, e.g.”
  • headersobject“Extra HTTP headers, e.g.”
  • actionsarray“Steps run after load to reach a state (modal, step 3)”
  • viewportobject
  • waitUntilstring“Navigation wait condition (default domcontentloaded)”
  • includeHtmlboolean“Include the rendered HTML (truncated)”
  • maxHtmlCharsinteger“HTML cap when includeHtml is set (default 20000)”
  • timeoutMsinteger“Overall inspection limit in ms (default 45000)”
• list-testson 1 other serverreads or searches

  “List the Playwright tests available in the project (file, line, full title, projects), with optional filtering.”

  Publisher declares: read-only yes · idempotent yes · open-world no

  5 inputs

  • projectRootstring“Project directory; defaults to the server working directory”
  • configstring“playwright.config path or 1-based index when the project has several configs”
  • testDirstring“Restrict scanning to this directory (inside the project)”
  • filterstring“Case-insensitive substring filter on "file › title"”
  • limitinteger“Max tests to return (default 500)”
• validate-selectoron 1 other serverreads or searches

  “Check a CSS selector, Playwright selector or locator (getByRole/getByTestId…, .first(), .filter()) on a live page: validity, match count (actions need exactly 1), sample matches and a sturdier verifi…”

  Publisher declares: read-only yes · idempotent yes · open-world yes

  8 inputs, 2 required

  • urlstringrequired“Full URL of the live page to test against (http/https)”
  • selectorstringrequired“Selector or locator, e.g.”
  • projectRootstring“Project whose Playwright launches the browser (default: server root)”
  • storageStatestring“Logged-in session: storageState JSON inside the project, e.g.”
  • headersobject“Extra HTTP headers, e.g.”
  • actionsarray“Steps run after load to reach a state (modal, step 3)”
  • viewportobject
  • timeoutMsinteger“Overall limit in ms (default 45000)”
• generate-e2e-teston 1 other servergenerates contentdescribed as changing data

  “Scaffold a Playwright test from a description using the project's REAL selectors.”

  Publisher declares: read-only no · destructive yes · idempotent no · open-world yes

  9 inputs, 1 required

  • descriptionstringrequired“What the test should cover, e.g.”
  • pageUrlstring“Page the test starts on (default: baseURL from playwright.config)”
  • projectRootstring“Project directory; defaults to the server working directory”
  • configstring“playwright.config path or 1-based index”
  • testDirstring“Directory for the generated spec (default: detected testDir or tests/)”
  • filestring“Exact spec path (*.spec.ts / *.test.ts) inside the project instead of the default location”
  • writeboolean“Write the file to disk (default true)”
  • overwriteboolean“Replace an existing spec at the target path (only specs this tool generated)”
  • liveInspectboolean“Verify discovered selectors against the live page (default true when a URL is known)”
• compare-visual-stateon 1 other servercreates or changes datadescribed as changing data

  “Visual regression check: screenshot a live page (or element), compare it to a stored baseline, and report how many pixels changed, WHERE they changed (bounding regions), and the average color shift (…”

  Publisher declares: read-only no · destructive yes · idempotent yes · open-world yes

  15 inputs, 2 required

  • urlstringrequired“Full URL of the page (or element target) to capture”
  • namestringrequired“Baseline id, e.g.”
  • projectRootstring“Project directory; defaults to the server working directory”
  • selectorstring“Capture only this element (selector or locator)”
  • fullPageboolean“Capture the full scrollable page (default false)”
  • waitUntilstring
  • waitForstring“Wait for this selector before capturing”
  • storageStatestring“Logged-in session: storageState JSON inside the project, e.g.”
  • headersobject“Extra HTTP headers, e.g.”
  • actionsarray“Steps run after load to reach a state (modal, step 3)”
  • viewportobject
  • timeoutMsinteger
  • actionstring“'baseline' overwrites the stored baseline; 'compare' (default) compares against it”
  • tolerancenumber“Percent of pixels that may differ before reporting a change (default 0.1)”
  • pixelThresholdinteger“Combined per-pixel channel delta considered different (default 60)”
• diagnose-flakyon 1 other serverruns code or commandsdescribed as changing data

  “Run a failing test multiple times with retries disabled and return an evidence-based verdict: CONSISTENTLY FAILING (real bug), FLAKY (intermittent — includes pass/fail counts and error-signature vari…”

  Publisher declares: read-only no · destructive yes · idempotent yes · open-world yes

  8 inputs

  • projectRootstring“Project directory; defaults to the server working directory”
  • testFilesarray“Tests to diagnose ("file:line" supported).”
  • runsinteger“How many times to run the tests (default 3)”
  • browserstring“Playwright project name to run”
  • headedboolean“Run with a visible browser window”
  • workersinteger
  • timeoutMsinteger“Hard wall-clock limit per run (default 120000)”
  • configstring“playwright.config path or 1-based index when the project has several configs”
• analyze-historyon 1 other serverreads or searches

  “Analyze the local run history (.playwright-e2e-mcp/history.jsonl, written by every run): 'flaky' ranks tests that flip pass/fail and lists broken or regressed ones; 'patterns' groups failures by caus…”

  Publisher declares: read-only yes · idempotent yes · open-world no

  5 inputs

  • viewstring“Default 'flaky'”
  • teststring“Filter: substring of file or title”
  • lastRunsinteger“Runs to analyze (default 30)”
  • limitinteger
  • projectRootstring

Each quoted sentence is the first sentence of the publisher’s own description. The kind of tool and the “changes data” mark are our AI model’s reading of that description, not something we tested: two differently worded runs had to agree, otherwise no label is shown, and nobody has checked them by hand. A tool without the “changes data” mark is described as only reading, searching or calculating, or the two runs did not agree. No tool was called. Read the publisher’s full description before relying on it. “Publisher declares” lists the hints the server itself sends with the tool (MCP annotations: read-only, destructive, idempotent, open-world); they are the publisher's claims, not checked by us. Inputs are the parameters in the tool's own input schema: name, declared type, whether it is required, and the first sentence of its description. Read 2026-10-23.

The registry verifies a package once, when it is published, and does not re-check it. A difference below is about the listing, not a judgement of the publisher.