Skip to main content
Glama
TestingBot MCP Server

TestingBot MCP Server

Official

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
TB_KEYNoAlias for TESTINGBOT_KEY.
LOG_LEVELNoSet to "debug" to enable debug logging.
TB_SECRETNoAlias for TESTINGBOT_SECRET.
TESTINGBOT_KEYNoYour TestingBot API key. Alias: TB_KEY or TESTINGBOT_USERNAME.
TESTINGBOT_SECRETNoYour TestingBot API secret. Alias: TB_SECRET or TESTINGBOT_ACCESS_KEY.
TESTINGBOT_PROFILENoSelects the credentials profile to use. Defaults to "default".default
TESTINGBOT_USERNAMENoAlias for TESTINGBOT_KEY.
TESTINGBOT_ACCESS_KEYNoAlias for TESTINGBOT_SECRET.
TESTINGBOT_CONFIG_DIRNoDirectory containing the TestingBot credentials file. Defaults to ~/.testingbot.
TESTINGBOT_TUNNEL_IDENTIFIERNoTunnel identifier to use for tb_openBrowser when a tunnel is needed.

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}
prompts
{}
resources
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
tb_loginA

Authenticate with TestingBot — no API key or secret to copy or paste. By default it opens your browser, you click Authorize, and the credentials are captured automatically and saved to ~/.testingbot/credentials (used by every later tool call this session — no restart). If your browser can't reach this machine (SSH / remote / dev container / web), it falls back to a short URL + code you enter in any browser; pass mode:"device" to force that. After authorizing in the browser, call tb_login again to finish. Call this whenever credentials are missing.

getBrowsersB

Get list of available browsers and platforms for testing. Optionally filter by type (web or mobile).

getDevicesB

Get list of available mobile devices for testing (real devices and simulators).

getTestsA

Retrieve a list of recent tests with optional pagination. Returns test details including status, browser, platform, video and duration.

getTestDetailsA

Get detailed information about a specific test by session ID. Includes logs, screenshots, video URLs, and execution metadata.

updateTestB

Update test metadata such as name, status (passed/failed), and other attributes. Useful for marking tests after execution.

deleteTestB

Delete a test by session ID. This permanently removes the test and its associated data.

stopTestB

Stop a running test by session ID. This terminates the test execution immediately.

getBuildsB

Get a list of builds with optional pagination. Builds group related tests together.

getTestsForBuildB

Get all tests associated with a specific build ID.

deleteBuildB

Delete a build and all its associated tests by build ID.

uploadFileA

Upload a local file (APK, IPA, or ZIP) to TestingBot storage for mobile app testing. Returns an app_url for use in tests.

uploadRemoteFileC

Upload a file from a remote URL to TestingBot storage. The file will be downloaded from the URL and stored.

getStorageFilesC

List all files in TestingBot storage with pagination.

deleteStorageFileC

Delete a file from TestingBot storage using its app_url.

takeScreenshotA

Take screenshots of a URL across multiple browsers and platforms. Returns a screenshot ID to retrieve results. localhost / private-network URLs are captured through the account's newest running TestingBot Tunnel — start one with startTunnel first.

retrieveScreenshotsA

Retrieve screenshot results by screenshot ID. Returns URLs to the generated screenshots.

getScreenshotListB

Get a list of all screenshot jobs with pagination.

getUserInfoA

Get current user account information including minutes used, plan details, and account status.

updateUserInfoC

Update user account information such as name, email, or other profile details.

startLiveSessionA

Start an interactive live testing session on TestingBot. Returns a link that opens the session in the user's TestingBot dashboard (they confirm with Start). Supports desktop browsers (Chrome, Firefox, Safari, Edge, IE) and mobile emulators/simulators (Android, iOS). For localhost / private-network URLs, start a tunnel first (startTunnel) and pass its numeric ID as tunnelId.

startDesktopLiveSessionB

Convenience tool to start a desktop browser live testing session. Automatically sets platformType to 'desktop'. For localhost / private-network URLs, pass the tunnelId of a running tunnel (startTunnel).

startMobileLiveSessionB

Convenience tool to start a live testing session on a mobile emulator/simulator. Automatically sets platformType to 'mobile'. For localhost / private-network URLs, pass the tunnelId of a running tunnel (startTunnel).

getTeamA

Retrieve team settings: plan details, team size, and concurrency (parallel sessions allowed, running now, and queued waiting for a free slot, for VMs and physical devices). Check this before starting many sessions in parallel.

getUsersInTeamA

Get a list of all users in your team with their roles and permissions.

getUserFromTeamC

Retrieve detailed information about a specific user in your team by their user ID.

createCdpSessionA

Create a remote browser session on TestingBot and get its CDP (Chrome DevTools Protocol) URL for direct browser control. Use this to automate browsers via CDP clients like Puppeteer or Playwright.

getTunnelListA

Get a list of all active TestingBot tunnels on the account, with their identifiers. Tunnels allow you to test websites behind firewalls or on your local machine.

deleteTunnelB

Delete an active TestingBot tunnel by its ID. This will terminate the tunnel connection.

startTunnelA

Start a TestingBot Tunnel on this machine so remote browsers and devices can reach localhost, private-network and intranet URLs. Call this BEFORE testing any local or private URL. Waits until the tunnel is ready (can take up to a minute on first use, which downloads the tunnel) and returns its identifier — pass that as tunnelIdentifier to tb_openBrowser / runMaestroTest. Requires Java 17+ on this machine. The tunnel keeps running until stopTunnel is called or the MCP server exits.

stopTunnelA

Stop a TestingBot Tunnel that was started with startTunnel. Without a tunnelIdentifier, stops the only one running (or all of them with all: true). For tunnels started outside this MCP server, use deleteTunnel.

getFailureLogsA

Fetch and return session log file content (selenium, browser, chrome, vm, appium) for a test by session ID. Use this to debug failures — pass failuresOnly=true to extract only error/exception/stack-trace lines.

getVisualTestsB

List the account's hosted visual regression tests (id, name, URL, browsers, paused).

getVisualRunsA

List visual runs (one browser's screenshot compared with its baseline), newest first. Use state=unreviewed to find failed runs nobody has rejected yet.

explainVisualRunA

Explain what changed in a visual run: each changed region with what it is, an AI description of the change when the account has AI region names on, text and CSS changes from the page, plus links to the screenshot, baseline and diff images. Read-only.

reviewVisualRunA

Approve or reject a visual run, the same as on its run page. Approving with makeBaseline=true also makes its screenshot the new baseline for that browser. Only do this when the user asked you to.

getAccessibilityTestsB

List the account's scheduled accessibility tests (each scans one URL) with their latest result.

getAccessibilityRunsA

List the scans of one accessibility test, newest first, with score and issue counts.

getAccessibilityFindingsA

Read the findings of an accessibility scan: failing WCAG A/AA rules with the elements (CSS selector, HTML, how to fix), checks that need human review, best-practice recommendations, and what is new or resolved since the previous scan.

runAccessibilityTestA

Start a scan of a scheduled accessibility test now (like Run now in the dashboard). Poll getAccessibilityRuns for the new run.

getAiFailureAnalysisA

Read the AI failure analysis (TestingBot AI Insights) of a test: summary, root cause, timeline, log evidence, suggested fixes, failure category, likely owner and confidence. Read-only: never starts an analysis or sends data to the AI provider. Requires the account owner to have enabled AI Analysis.

analyzeTestFailureA

Explain why a test failed with TestingBot AI Insights. Starts an AI failure analysis of a finished test (or returns the existing one without extra cost) and waits for the result. The test's curated, secret-masked logs are sent to TestingBot's AI provider, so this requires the account owner to have enabled AI Analysis in Account Settings. Use getFailureLogs instead when you want the raw logs.

listTestFilesB

Scan a project directory and return test files for the detected framework. Read-only — never modifies files.

setupTestingBotA

Detect the testing framework in a project and return the ready-to-paste TestingBot configuration snippet, install command, and env-var requirements. Read-only — never modifies files; the agent or user applies the snippet.

maestroCheatSheetA

Get a Maestro flow YAML syntax reference: file structure, commands, selectors, and the TestingBot run workflow. Call this before writing or fixing Maestro flows.

uploadMaestroAppA

Upload a mobile app (APK, AAB, IPA, or zipped .app) for Maestro testing on TestingBot. Skips the upload if the same binary was uploaded before. Returns a projectId used by uploadMaestroFlows and runMaestroTest.

uploadMaestroFlowsA

Upload Maestro flow YAML files for a project. Accepts a local directory (all .yaml/.yml files, preserving structure), a pre-built .zip, or inline flow content. Replaces any previously uploaded flows for the project.

runMaestroTestA

Start a Maestro test run on a TestingBot device. Returns immediately with the run ID — it does NOT wait for completion. Poll with getMaestroRunStatus, then fetch getMaestroRunResults once completed.

getMaestroRunStatusA

Get the status of Maestro run(s) for a project. Omit runId to list all runs of the latest execution. Statuses: WAITING, READY (running), DONE, FAILED, CANCELLED.

getMaestroRunResultsA

Get the final results of a completed Maestro run: per-flow outcomes (failures first, with error messages and video links) plus the JUnit XML report.

uploadMaestroCompanionAppA

Upload a companion app (.apk/.aab/.ipa) that should be installed alongside the app under test in a Maestro run. Returns a tb:// app URL to pass in runMaestroTest's otherApps parameter (max 4 per run).

listMaestroProjectsA

List Maestro projects on the account (newest first): app details, flow names, and run IDs. Use this to find an existing projectId instead of re-uploading an app.

listMaestroRunsA

List Maestro runs across all projects (newest first), or look up runs by their build/run name. Returns projectId + runId pairs for use with getMaestroRunStatus and getMaestroRunResults.

getMaestroFlowDetailsA

Get the detailed result of a single Maestro flow: error messages, step-level JUnit report, and links to the video, screenshots, and device logs. Use this to diagnose why a flow failed before fixing it and calling retryMaestroRun.

cancelMaestroRunC

Cancel a running Maestro run.

retryMaestroRunB

Retry a Maestro run, or a single failed flow within it when flowId is given.

uploadAppAutomateAppA

Upload the app under test for an Espresso (.apk/.aab) or XCUITest (.ipa) run on TestingBot. Returns a projectId for uploadAppAutomateTests and runAppAutomateTest. For Maestro flows use uploadMaestroApp instead.

uploadAppAutomateTestsB

Upload the test suite for an App Automate project: the instrumented test APK for Espresso, or the zipped XCUITest runner bundle for XCUITest.

runAppAutomateTestA

Start an Espresso or XCUITest run on a TestingBot device. Returns immediately with the run ID — it does NOT wait for completion. Poll with getAppAutomateRunStatus, then fetch getAppAutomateRunResults. If the app needs a localhost / private-network backend, start a tunnel first (startTunnel): the run uses the account's newest running tunnel. On Android, point the app at a hostname that resolves to 127.0.0.1 on the tunnel's machine (e.g. myapp.localtest.me), not localhost, which Android never proxies.

getAppAutomateRunStatusA

Get the status of Espresso/XCUITest run(s) for a project. Omit runId to list all runs. Statuses: WAITING, READY (running), DONE, FAILED.

getAppAutomateRunResultsB

Get the results of a completed Espresso/XCUITest project: per-run outcomes plus the JUnit XML report. For XCUITest, pass runId for a single run's report.

tb_openBrowserA

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.

tb_navigateA

Navigate the session's browser to a URL. By default it waits for the page to finish loading; use waitUntil to return earlier on a slow page. localhost / private-network URLs only load when the session was opened with tunnelIdentifier (see tb_openBrowser).

tb_snapshotA

Return a structured snapshot of the current page: title, URL, headings, links/buttons with their text, and a truncated body. Use this as the agent's 'see the page' primitive before deciding what to click.

tb_clickC

Click an element. Use by to pick a locator strategy (css/xpath/id/name/tag/class/linkText/partialLinkText).

tb_typeA

Type text into an input. Uses the same locator strategies as tb_click. Clears the field first by default.

tb_getTextA

Read the visible text of an element. Cheaper than tb_snapshot when you just need to verify a single label / status.

tb_getAttributeB

Read an attribute of an element (e.g. 'href', 'value', 'aria-label', 'class', 'data-*').

tb_executeScriptB

Execute a JavaScript snippet in the active page and return its result. The script body is wrapped as a function — use return to send a value back. Args are passed positionally via arguments[0], arguments[1], ...

tb_closeBrowserA

Close a remote browser session. Releases TestingBot minutes. For mobile device sessions, use appium-mcp's appium_session_management with action: 'delete'.

tb_screenshotA

Capture a PNG screenshot of the active browser session. Returns a base64 image content block AND saves the PNG to disk, returning a clickable file:// link — many MCP clients (e.g. Claude Desktop) don't render tool-returned images inline, so the saved file is the reliable way for a human to view it. Pass savePath to control where it's written (a .png file path, or a directory); otherwise it goes to the OS temp dir.

tb_pressKeyA

Press a key (or chord) on the active browser session. Uses W3C key names: 'Enter', 'Tab', 'Escape', 'Control+A', 'Meta+Shift+Z', etc. Sent to the currently-focused element.

tb_listSessionsA

List all active browser sessions managed by this server. Useful when you've lost track of a sessionId. (Mobile device sessions are NOT listed here — they live inside appium-mcp; use appium_session_management with action 'list'.)

Prompts

Interactive templates invoked by user choice

NameDescription
why-did-my-test-failInvestigate one failed TestingBot test: status, AI failure analysis when enabled, and the failing log lines.
summarise-buildSummarise a TestingBot build: pass/fail counts, the failures grouped by cause, and what to fix first.
find-flaky-testsFind tests that both passed and failed recently and suggest why they are unstable.

Resources

Contextual data attached and managed by the client

NameDescription
flaky-testsTests that both passed and failed in the last 30 days, most flaky first. Add ?time_range=7d to change the window.

TDQS

B3.2/5.0

Scored across 73 tools

Disambiguation3/5

Most tools target distinct resources or workflows, and descriptions often clarify family-specific tools (Maestro vs App Automate vs visual vs accessibility). However, there is notable overlap among session-start tools (startLiveSession, startDesktopLiveSession, startMobileLiveSession, tb_openBrowser, createCdpSession), tunnel cleanup tools (stopTunnel vs deleteTunnel), and screenshot-related tools. The large surface makes boundary decisions harder even when each description is informative.

Naming Consistency3/5

Many tools follow a camelCase verb-noun convention, but the set mixes unprefixed camelCase names, a tb_ snake_case-prefixed browser-control family, and noun-style names like maestroCheatSheet. The tb_ grouping is deliberate and readable, but overall naming is not uniform across the server.

Tool Count2/5

At 73 tools, the server is far beyond the typical 3-15 well-scoped range and is heavy even for a broad testing platform. Several tools are convenience wrappers or narrow variants, which adds surface area without proportional independent capability.

Completeness3/5

The server covers many TestingBot workflows: sessions, tunnels, storage, tests/builds, Maestro, App Automate, accessibility, visual regression, account, and team. Still, several subdomains lack full lifecycle coverage, such as create/update/delete for accessibility and visual tests, cancelling App Automate runs, and deleting Maestro projects. These are notable gaps but agents can often work around them through the dashboard or existing tools.

Maintenance

ActivityMaintained
ResponsivenessNo recent community issues