# Katalon TestOps MCP for AI Agents AI Agent Connect

> Katalon TestOps gives your AI agent direct access to your test management platform. You can trigger test reruns, check execution statuses, and pull diagnostic logs without leaving your workspace. It connects your automation suite to your AI client so you can monitor release quality and debug failures in real time.

## Overview
- **Category:** ship-it
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_ovZCFfCK6Kz1sdVzEWtQhoPAzq5RaP9rqfTzVwdi/ai-agent-connect
- **Tags:** test-orchestration, quality-assurance, test-automation, software-testing, ci-cd-pipeline, test-analytics

## Description

Imagine you just pushed a major update and your CI/CD pipeline is screaming. Instead of logging into a separate dashboard, flipping through tabs, and hunting for the specific test run that failed, you just ask your AI agent what is going wrong. This Connector lets your agent do the heavy lifting. It can pull the exact error stack traces, grab screenshots of the UI where the test broke, and tell you which environments are actually running the code. You can tell your agent to rerun a specific suite of tests to see if your latest fix actually worked, or ask for a summary of the pass rates across your entire project. It turns your test management into a conversation. By adding this to your setup via Vinkius, you stop acting as a middleman between your code and your test results. You get to spend more time fixing bugs and less time hunting for them. You can check which browsers or operating systems are being used for specific test runs to ensure your target audience is actually being covered. When a release is about to go live, you can get a high-level summary of the test statistics to confirm everything is green. This means you stay in your flow without the constant context switching that usually comes with software testing. You get clear answers on build quality and execution status right where you are already working.

## Tools

### get_test_result
Pull the full details for a single test result. Use this to see specific error messages or logs for one test.

### list_execution_environments
See which OS, browser, and device targets are configured. This helps you verify where your tests are actually running.

### list_project_releases
View scheduled releases and their aggregated test statistics. This is great for checking release readiness.

### list_projects
Get a list of all projects in your Katalon TestOps account. Use this to see every project you manage.

### get_project
Pull all the details for a specific project. This gives you the full context for a single project.

### list_test_runs
See the status, duration, and pass/fail counts for all runs in a project. It gives a quick overview of run history.

### get_test_run
View the complete details of a specific test run. This is useful for deep-diving into a single execution.

### rerun_test_run
Start a new execution for an existing test run. Use this to quickly retry tests that failed on the first try.

### list_test_results
Get a list of every test result with error messages and durations. This helps you see which specific tests failed.

### list_project_builds
See all software builds in a Katalon project. This shows you the build IDs and how many test runs they contain.

## Prompt Examples

**Prompt:** 
```
List the last 5 test runs in project 'E-commerce API' (ID: 123)
```

**Response:** 
```
I've retrieved the latest 5 test runs for project 123. 

| Run ID | Status | Duration | 
| :--- | :--- | :--- | 
| #987 | ✅ Passed | 45s | 
| #986 | ❌ Failed | 12s | 
| #985 | ✅ Passed | 48s | 
| #984 | ✅ Passed | 42s | 
| #983 | ✅ Passed | 44s | 

Would you like me to inspect the results for the failed run #986?
```

**Prompt:** 
```
Rerun test run ID 986
```

**Response:** 
```
Rerunning test run 986… 

**New active execution started.** 
New Run ID: 988. 

I'll let you know once the results are available. Would you like me to notify you if it fails again?
```

**Prompt:** 
```
What are the active releases defined in project 123?
```

**Response:** 
```
I've identified 2 active releases in project 123:

*   **Q1 Spring Update** (v1.2.0) - 45 runs, 98% pass rate
*   **Hotfix Security** (v1.2.1) - 12 runs, 100% pass rate

Would you like a detailed build report for either of these?
```

## Capabilities

### Rerun failed test suites
Trigger new executions for specific test runs directly from your chat.

### Pull error stack traces
Get the exact logs and error messages for failed tests instantly.

### Check release pass rates
See aggregated test statistics for your scheduled software releases.

### Audit execution environments
View the OS, browser, and device targets for your test projects.

### Fetch project metadata
Retrieve project IDs, descriptions, and team assignments.

### List software builds
See all builds in a project and the number of test runs they contain.

## Use Cases

### Debugging a flaky test
An engineer asks the agent to pull the screenshot and stack trace for a specific failed run to see why it broke.

### Release sign-off
A QA manager asks for the pass rate of the Q1 Update release to see if the build is ready for production.

### Rapid iteration
A dev pushes a hotfix and asks the agent to rerun only the failed suite from the previous build to verify the fix.

### Environment audit
A DevOps lead asks which browsers are currently supported in the E-commerce project to verify configuration.

## Benefits

- Quickly identify bugs by using `get_test_result` to pull error messages and logs into your chat.
- Save time on manual reruns by using `rerun_test_run` to trigger new executions instantly.
- Monitor project health across multiple teams using `list_projects` and `list_test_runs`.
- Ensure environment parity by checking targets with `list_execution_environments`.
- Track release readiness with `list_project_releases` to see pass rates before deployment.
- Audit build history using `list_project_builds` to see associated test counts.

## How It Works

The bottom line is you get a voice-controlled interface for your entire test orchestration pipeline.

1. Subscribe to the Connector on Vinkius.
2. Provide your Katalon Email and API Key in the configuration.
3. Ask your AI agent to check test results or trigger new runs.

## Frequently Asked Questions

**Can Katalon TestOps help me find why a test failed?**
Yes. It allows your AI agent to pull the exact error messages and logs for a specific test result so you can see the cause of the failure without leaving your chat.

**Can I trigger test reruns with Katalon TestOps?**
Yes, you can ask your agent to rerun any existing test run. It will trigger the new execution and provide you with the new run ID.

**Does Katalon TestOps support multiple projects?**
Yes. You can list all projects in your account and then fetch specific details, runs, or releases for any of them.

**How does Katalon TestOps help with release management?**
It provides a way to view all defined releases in a project along with their aggregated test statistics, helping you verify release readiness quickly.

**Can I check which browsers my tests run on?**
Yes, the Connector can list the execution environments for your projects, including the target operating systems, browsers, and device distributions.

**Can I use Katalon TestOps with Claude or Cursor?**
Yes, this Connector is compatible with any MCP-supported client, including Claude, Cursor, and Windsurf, making it easy to add to your existing workflow.

**Can I rerun a failed test suite using my agent?**
Yes. Use the `rerun_test_run` tool by providing the ID of the failed execution. Your agent will trigger a fresh run using the exact same configuration and return the new Test Run ID for tracking.

**How do I see the screenshots from a specific test failure?**
Ask your agent to `get_test_result` for a specific result ID. If the test was configured to capture visuals, your agent will retrieve the screenshot links along with diagnostic logs and error messages.

**Can I check which browser versions were used in a test run?**
Absolutely. Use the `list_execution_environments` tool for your project. Your agent will return detailed OS, browser, and device configurations used for executions, ensuring you have the full technical context.