Skip to main content
Glama
TestingBot MCP Server

TestingBot MCP Server

Official

tb_openBrowser

Start a remote WebDriver browser session on TestingBot for desktop or mobile web pages. Returns a sessionId and live view URL; pass a tunnel identifier to reach localhost or private sites.

Instructions

Start a remote browser session on TestingBot via WebDriver. Handles BOTH desktop browsers (Chrome/Edge/Safari/Firefox on Win/macOS/Linux) AND mobile browsers (Chrome on Android, Safari on iOS). For a mobile browser task, set deviceName and platformVersion — the session is still WebDriver/chromedriver-backed, so tb_navigate / tb_snapshot / tb_screenshot / tb_click work directly with no native-context juggling. By default a mobile session runs on an EMULATOR (Android) / SIMULATOR (iOS); set realDevice: true to run on a physical device instead — ALWAYS do this when the user asks for a 'real device', 'physical device', or 'real hardware'. Prefer this over appium_session_management for any task whose goal is loading a web page in a browser, even on mobile. Reserve appium_session_management for native-app testing (.apk/.ipa). Local or private sites (localhost, 127.0.0.1, 10.x/192.168.x, intranet hosts): the remote browser can only reach these through a running TestingBot Tunnel, and only when this session names that tunnel via tunnelIdentifier — without it the session silently goes direct and the page fails to load. Returns a sessionId for subsequent tool calls and a liveViewUrl the human can open to watch the agent drive in real time. Burns TestingBot minutes — call tb_closeBrowser when done.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "browserName": "<browserName>",
      "platform": "<platform>"
    },
    "name": "tb_openBrowser"
  }
}

Input Schema

NameRequiredDescriptionDefault
nameNoHuman-readable session name (shown in dashboard)
buildNoBuild identifier for grouping
platformYesPlatform/OS. Desktop codes like 'WIN11', 'VENTURA', 'macOS Ventura', or for mobile pass 'Android' or 'iOS' (and also set deviceName + platformVersion).
deviceNameNoMobile device name (e.g. 'Google Pixel 8', 'iPhone 15 Pro'). REQUIRED for mobile-browser sessions. Triggers Appium-style capabilities under the hood, but the resulting session is still WebDriver — use tb_navigate / tb_snapshot etc. as usual.
realDeviceNoRun on a physical device instead of an emulator/simulator (mobile only). Set true whenever the user asks for a 'real device' or 'physical device'. Ignored for desktop. If the requested physical device is unavailable, this tool returns (without starting a session) a list of available alternatives plus the option to wait or use an emulator — relay those choices to the user.
browserNameYesBrowser engine
automationNameNoMobile automation engine override. Defaults: 'UiAutomator2' for Android, 'XCUITest' for iOS. Only set this if you have a specific reason.
browserVersionNoBrowser version (desktop only, default: 'latest'). Ignored for mobile.latest
platformVersionNoMobile OS version (e.g. '14' for Android 14, '17' for iOS 17). Pair with deviceName.
screenResolutionNoScreen resolution for desktop sessions (e.g., '1920x1080'). Ignored for mobile.
tunnelIdentifierNoIdentifier of a running TestingBot Tunnel (the value it was started with via --tunnel-identifier, or the one startTunnel returned) to route this session through. REQUIRED to open localhost / private-network URLs. Defaults to the TESTINGBOT_TUNNEL_IDENTIFIER environment variable when set.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv1.3.0

TDQS

A4.8/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations supplied, the description carries the full behavioral burden and does so well. It discloses default emulator/simulator behavior, the consequence of omitting tunnelIdentifier for private sites (the session silently goes direct and the page fails), the fact that sessions consume TestingBot minutes, the return values (sessionId and liveViewUrl), and the alternative-device list returned instead of a session when a requested physical device is unavailable.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is long but front-loaded and organized by decision: purpose, mobile support, real-device selection, sibling alternative, tunnel handling, returns, and cost. Some phrasing repeats schema details, particularly around deviceName/platformVersion and realDevice, but the additional length is largely justified by the tool's complexity.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For an 11-parameter session-creation tool with no annotations and no output schema, the description covers the essential operational context: how to start desktop vs. mobile sessions, how real devices differ, when tunneling is required, what is returned, and the cost/cleanup expectation. An agent has enough information to invoke it correctly in the common cases.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so parameter documentation is already strong and the baseline is 3. The description adds value beyond the schema by explaining the operational consequence of tunnelIdentifier and by reinforcing the realDevice requirement in task terms. It does not explain every parameter, but the additional semantics it supplies are meaningful.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb and resource — 'Start a remote browser session on TestingBot via WebDriver' — and immediately scopes it to both desktop and mobile browsers. It distinguishes the tool from appium_session_management by stating that web-page tasks should use this tool while native-app testing should not. An agent can identify the tool's role without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The guidance is unusually explicit: it names the alternative (appium_session_management) and the condition that selects it, explains how to handle mobile sessions, and states precisely when to set realDevice to true. It also warns that local/private sites require tunnelIdentifier and says to call tb_closeBrowser when done. There is little left to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.