# Mabl MCP

> Mabl MCP gives your AI client full control over complex E2E testing and quality orchestration. You can list all monitored applications and environments, trigger test plans across staging or production, monitor live runs, and deep-dive into failure details to pinpoint bugs instantly. It turns manual QA checklists into conversational commands.

## Overview
- **Category:** developer-tools
- **Price:** Free
- **Tags:** test-automation, e2e-testing, quality-assurance, low-code, failure-analysis, ci-cd-integration

## Description

This connector lets you treat your entire test suite like a conversation with your agent. Instead of navigating through complex UIs, you simply tell it what needs testing. You can list all available applications and then trigger specific test plans—say, the 'Checkout Regression' plan on the staging environment. Your agent monitors the run in real time, providing pass/fail rates and duration summaries as it progresses. If something breaks, you don't just get a red status; you get detailed failure analysis. The system extracts precise error reasons, diagnostic insights, and even visual screenshots for rapid debugging. This capability is part of Vinkius’s massive catalog, giving your AI client deep control over your quality checks directly from your workspace.

## Tools

### mb.list_apps
Lists all applications monitored by Mabl, including details for web, mobile, and API testing.

### mb.get_app
Retrieves full information on a specific Mabl application, listing its targets (web URLs plus API base URLs) and labels.

### mb.list_envs
Lists all testing environments available in Mabl, returning their names, IDs, and defined variables.

### mb.list_plans
Returns a list of all test plans (run policies), showing their status, triggers, and linked applications or environments.

### mb.trigger_plan
Starts an immediate run of a specific Mabl plan via deployment event trigger.

### mb.list_executions
Lists the most recent test execution results, giving a quick overview of outcomes.

### mb.get_execution
Retrieves comprehensive details for one specific test run, including status, duration, and failure analysis reports.

### mb.list_tests
Lists all available API tests configured within Mabl.

### mb.list_labels
Returns a list of organizational labels (tags) used to group and manage test plans.

### mb.workspace_info
Gets general metadata about the Mabl workspace, including organization details.

## Prompt Examples

**Prompt:** 
```
List all my Mabl test plans
```

**Response:** 
```
I've found 4 test plans: 'Daily-Smoke-Test' (ID: plan-123), 'Checkout-Regression' (ID: plan-456), 'Login-Flow-Validation' (ID: plan-789), and 'API-Health-Check' (ID: plan-012). Which one would you like to trigger or inspect?
```

**Prompt:** 
```
Trigger the 'Checkout-Regression' plan in the Staging environment
```

**Response:** 
```
Triggering test plan 'Checkout-Regression'… I've successfully started the run in your Staging environment. I'll monitor the progress and notify you once the execution is complete. Would you like the initial status report?
```

**Prompt:** 
```
Show me the details for the last failed execution
```

**Response:** 
```
Retrieving the most recent failed execution… Run ID 'exec-98765' failed 12 minutes ago during the 'Add to Cart' step. The AI analysis suggests a 'Button not clickable' error due to a visual overlay. I can provide the direct screenshot link if you'd like.
```

## Capabilities

### Identify all applications
Retrieves a list of every web and API application monitored by Mabl.

### Manage test environments
Lists configured testing environments along with their associated variables for context-aware testing.

### Execute test plans
Triggers a full, automated run of any defined test plan across specific deployment stages.

### Monitor live runs
Tracks the real-time status and retrieves detailed outcome summaries for ongoing test executions.

### Analyze failure reports
Deep-dives into failed test runs to extract exact failure reasons, diagnostic text, and screenshots.

### Discover plan labels
Lists the organizational tags used to group and select specific testing suites.

## Use Cases

### Finding the source of an intermittent bug
A developer notices a UI regression but can't reproduce it locally. They ask their agent to find all applications using `mb.list_apps`, select the target app, and then trigger a run on the staging environment using `mb.trigger_plan`. The resulting failure analysis provides screenshots showing the exact visual overlay causing the button click failure.

### Auditing a release candidate build
A DevOps engineer needs to confirm that all critical components passed testing before deployment. They ask their agent to list all test plans using `mb.list_plans`, select the 'Smoke Test' plan, and then use `mb.get_execution` on the last run ID to verify the overall pass rate.

### Validating API changes in a new environment
A backend team updates an API endpoint and needs QA to validate it quickly. They first list all environments using `mb.list_envs` to confirm the correct staging URL, then use `mb.trigger_plan` on the dedicated 'API Health Check' plan.

### Reviewing test coverage gaps
A QA engineer wants to know what tests are grouped together. They list all available labels using `mb.list_labels` and then use `mb.get_app` for the main checkout app to confirm which specific API endpoints were monitored.

## Benefits

- Stop manually cross-referencing build logs and test dashboards. By using `mb.get_execution`, you get a single, comprehensive report that includes pass/fail rates, duration, and detailed failure analysis.
- Never guess which environment to target again. Run `mb.list_envs` first, then tell your agent exactly where the plan needs to execute—staging or production—ensuring the right variables are loaded.
- Reduce debugging time from hours to minutes. The MCP can extract precise root causes and visual screenshots for failed tests using the results of a run, bypassing vague error messages.
- Manage complexity by organizing your suite. Use `mb.list_labels` to see all available tags and then trigger specific groups of tests using the `mb.trigger_plan` tool.
- Keep track of what's running without switching tabs. You can use `mb.list_executions` for a quick status check, or `mb.get_execution` for a deep dive into any single test run history.

## How It Works

The bottom line is you get full, conversational control over complex E2E testing workflows without opening a browser tab.

1. Subscribe to this MCP and enter your Mabl API Key into your agent.
2. Directly ask your AI client to list applications or environments it can manage.
3. Tell the agent which test plan to run, what environment to use, and wait for the failure analysis report.

## Frequently Asked Questions

**How do I list all my Mabl applications using the Mabl MCP?**
You use `mb.list_apps`. This tool gives you a complete catalog of every web, mobile, and API application that is currently monitored by your account.

**Can I run a test plan on different environments using mb.trigger_plan?**
Yes, `mb.trigger_plan` allows you to execute specific plans across defined deployment stages like staging or production, ensuring the correct context for your tests.

**What information does mb.get_execution provide about a failed test run?**
It gives full details of the execution, including the status, duration, pass/fail rates, and critical failure analysis reports with diagnostic insights and screenshots.

**How do I check what environments are available for testing?**
You run `mb.list_envs`. This function returns all defined environment names, unique IDs, and any associated variables needed by the test plan.

**Does the Mabl MCP help me find out which tests belong to a specific group?**
Yes, you can use `mb.list_labels` to view all organizational labels (tags). You can then reference these labels when triggering test plans to select grouped suites.