# PractiTest MCP for AI Agents AI Agent Connect > PractiTest connects your QA management workspace to your AI agent. It lets you list tests, trace requirements to defects, and pull execution histories directly into your chat. Stop jumping between dashboard views to find out why a payment flow failed; just ask your agent to find the specific test run and the linked issue. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_aImDfxsnDktkH2b8NuISoxnqQMdLW9rlwg57urZq/ai-agent-connect - **Tags:** qa-testing, test-management, bug-tracking, quality-assurance, test-automation ## Description Connect your PractiTest workspaces to any AI agent and let it handle the heavy lifting of your quality assurance workflow. Instead of jumping between different dashboards to see if a requirement is met or why a specific test failed, you can just ask your agent to find the information for you. It can pull your test sets, look into specific execution histories, and map software defects back to the original requirements. This turns your QA data into a searchable knowledge base where you can trace the entire path from a high-level goal down to a specific bug. You'll spend less time hunting through tabs and more time actually solving problems. Because this is hosted on Vinkius, it's easy to plug into your existing AI setup and get started immediately. ## Tools ### create_run Provide the data as a JSON string. Create a new run in a PractiTest project ### create_test Provide the data as a JSON string. Create a new test in a PractiTest project ### get_project Get details of a specific PractiTest project ### get_requirement Get details of a specific requirement in a PractiTest project ### list_requirements View all project requirements and see how many tests are linked to each. ### create_instance Provide the data as a JSON string. Create a new instance in a PractiTest project ### list_instances Find all test instances within a set, including their last run statuses. ### list_projects List all PractiTest projects accessible by the API token ### list_runs Pull the full history of execution results, durations, and timestamps for a test instance. ### list_tests Pull all test cases in a project with traceability links and custom fields. ### get_test Get the full details of a test case, including steps, preconditions, and requirements. ## Prompt Examples **Prompt:** ``` Can you show me the status of the 'User Auth' test set and tell me who's assigned to it? ``` **Response:** ``` I've pulled the details for the **User Auth** test set: | Test Set | Status | Assigned Testers | | :--- | :--- | :--- | | User Auth | **IN_PROGRESS** | Sarah Jenkins, Mike Ross | Would you like me to pull the specific instances for this set? ``` **Prompt:** ``` Find the highest severity bugs that are linked to failed tests in our regression suite. ``` **Response:** ``` I've scanned the issues for you. Here are the high-priority defects linked to failed tests: * **Issue #91**: EUR vs GBP Cart Rounding Error (Severity: **High**) * **Issue #104**: Checkout Button Unresponsive (Severity: **Critical**) Both are currently marked as **UNRESOLVED**. Should I get the test history for either of these? ``` **Prompt:** ``` What are the preconditions for the 'Multi-currency Checkout' test case? ``` **Response:** ``` Here are the preconditions for the **Multi-currency Checkout** test case: * User must have a verified account. * Currency must be set to a non-USD value. * The cart must contain at least one item with a price greater than 0.00. I can also pull the expected results if you'd like. ``` ## Capabilities ### Trace requirements to tests Map high-level requirements directly to the test cases that verify them. ### Pull execution histories Retrieve full logs of test runs to see exactly what happened during a cycle. ### Identify failed test steps Pinpoint the specific step in a test run that caused a regression. ### List software defects Get a list of all issues in a project along with their severity and links. ### Check custom field values View all custom metadata across different entities in your project. ### Get test preconditions Retrieve the exact setup needed before running a specific test case. ## Use Cases ### Regression Analysis A QA engineer asks the agent to find all failed runs in the Payment Flow set and list the linked issues using `list_runs` and `list_issues`. ### Requirement Auditing A Product Owner wants to know which requirements are missing test coverage and asks the agent to list requirements with zero links. ### Defect Triage A developer asks the agent to find the specific test steps that failed for a high-severity bug to see the expected behavior using `get_test`. ### Project Onboarding A new team member asks the agent to list all active test sets and the users assigned to them to get a project overview using `list_sets` and `list_users`. ## Benefits - Stop manual tracing: Link requirements to tests and issues instantly using `list_requirements` and `list_issues`. - Faster root cause analysis: Use `list_runs` to see exactly which step caused a regression without manual searching. - Real-time status checks: Get live updates on test sets and requirements directly in your chat using `list_sets`. - Better data visibility: Access all custom fields across your project to see specific metadata using `list_custom_fields`. - Instant test discovery: Find any test case or set quickly to see preconditions and results using `list_tests` and `list_sets`. ## How It Works The bottom line is you get a direct line from your AI agent to your entire QA lifecycle data without leaving your chat window. 1. Get your Personal API Token and Project ID from your PractiTest account. 2. Add these credentials to your Connector client settings. 3. Ask your agent to pull test data, list requirements, or map issues in plain English. ## Frequently Asked Questions **Does the PractiTest MCP help with requirements traceability?** Yes, it connects your requirements directly to your tests and issues. You can ask your agent to show you exactly which requirements are covered by which tests. **Can I use PractiTest to find out why a test failed?** Yes, the Connector pulls run histories so you can see the exact steps that were executed. This helps you identify the specific point where a regression occurred. **How do I get my requirements into my AI agent?** The Connector pulls them directly from your PractiTest project. Once connected, your agent can read and summarize your requirements in any chat window. **Can my AI agent see custom fields in PractiTest?** Yes, it can list and read all your custom fields. This means your agent can access specific metadata that your team uses to categorize tests or issues. **Is this good for QA automation engineers?** It's a great fit. It helps engineers verify integration outputs and quickly find failed runs across large test suites without manual searching. **Can the AI provide the exact step where a test case failed?** Yes. If an execution failed, the agent uses `list_runs` for the instance. Since an instance maps directly to test steps, the AI inherently decodes the exact execution traces to show you the failing parameters. **Is PractiTest's requirement and issue tracing accessible to the AI?** Yes. Tools like `list_requirements` and `list_issues` expose full traceability trees. You can ask exactly how many QA instances are mapped to Requirement 5. **Do I need to copy the project ID separately?** Yes. In PractiTest, APIs execute cleanly isolated within specific Project instances. You must provide the numeric Project ID alongside your Personal Token so the underlying `pt-engine` binds queries strictly to that project. **Can the AI Agent execute tests inside PractiTest?** While the agent cannot run automated testing scripts directly in PractiTest, it can create Test Runs, log results into Instances, and manage the administrative side of QA efficiently. **Are custom fields supported when creating new tests?** Yes! The AI agent formats API requests dynamically. If your workspace requires custom fields, simply instruct the agent on which attributes to include during the test creation. **Is there a limit on how many tests the agent can list at once?** The agent adheres to PractiTest API pagination limits. By default, it returns a single page of results, but you can explicitly ask the AI to query a different page number or limit.