Quick Start Examples
Core Workflows
Creating Tests
Ask the assistant to generate tests for specific flows or features. It will suggest test cases that you can review, edit, or accept.“Generate 5 tests for the checkout flow”
“Create a test that validates user registration with email verification”You can also provide the assistant with a set of test cases (as text, a document, or by referencing existing tests) and have it create new tests or update current ones based on those examples.
“Create tests for the returns flow similar to my checkout tests”
“Here are my test cases: [paste list]. Generate tests based on these patterns or update current ones”The assistant analyzes your application’s knowledge graph and any examples you provide to suggest relevant tests. Tests appear as interactive suggestions - click “Add Selected Tests” to commit them to your project.
Running Tests
Ask the assistant to run tests and you’ll be notified when they complete.“Run the login test”
“Execute all checkout tests against the staging environment”What happens: The assistant validates your test cases, starts execution in the background, and you’ll see results when the run completes. Test runs typically take a few minutes depending on complexity.
Accessing Test Run Results
The assistant can access detailed step-by-step results for test runs, but with an important limitation: What the assistant CAN access:- Detailed step-by-step results (screenshots, actions, step descriptions) for test runs it triggered in the current chat conversation
- Action logs showing what happened during test execution
- Visual evidence of test steps and outcomes
- Runs triggered from the UI dashboard
- Runs triggered via API
- Scheduled test plan runs
- Runs from previous chat conversations
- Runs by run ID or test case ID (only runs linked to the current conversation)
Editing Existing Tests
Describe changes to tests in natural language. The assistant shows a visual diff for you to review before applying.“Change step 3 to wait for the spinner first”
“Add verification that the success message appears”
“Make this test resume from my Login test”After you approve changes, the test runs automatically to validate the updates. See Creating Tests for more editing examples.
Analyzing Your Application
Ask the assistant to crawl and analyze your site to expand its knowledge of your application’s features.“Crawl my admin panel”
“Analyze the checkout flow to find features I should test”The assistant can start from a specific URL or resume from a test’s output state (useful for authenticated areas). Results include screenshots and discovered interactions. See Crawling Sessions for configuration details.
Make the Assistant More Effective
The assistant understands context and can handle follow-up questions, so you can have natural back-and-forth conversations about your testing needs. It relies on Crawling, tests, and Knowledge for context gathering.Add Knowledge to improve test creation: The Chat Assistant works best when it understands your application. Add documentation URLs, project context, and domain-specific information to your Knowledge Base so the assistant can provide more accurate test suggestions and better understand your unique workflows.The assistant automatically searches your knowledge base and knowledge graph when generating tests and answering questions, giving you context-aware responses tailored to your specific application.Note: Knowledge and the knowledge graph are used when creating tests through chat, but not during test execution. When tests run, the agent only uses the test’s goal, steps, and a targeted subset of knowledge - Agent Rules, Agent Visuals, the Project Summary, and any knowledge items attached to the test - to navigate and verify results.
Uploading Files and Documents
You can upload files directly into chat to give the assistant additional context for test creation. Supported Formats:
PDF Processing:
When you upload a PDF, the AI reads both the text content and any images, diagrams, or screenshots within the document. This means you can upload:
- Product specifications with annotated screenshots
- Design documents with wireframes
- Requirements documents with flowcharts
- Bug reports with visual evidence
- Files remain available throughout your chat conversation
- You can reference uploaded content in follow-up messages
- Very long conversations may compress older file content to stay within limits, but recent uploads are always fully available
Files uploaded in chat are isolated to that conversation - other team members cannot see them. For shared documentation, use Project Knowledge instead.
What Can You Do with the Chat Assistant
Test Management The assistant can list your existing tests, generate new test cases based on your requirements, run tests in the background, and fetch detailed information about specific tests including their steps, configurations, and dependencies. It understands test scenarios (grouping folders) and test plans, making it easy to organize and trigger your test suite. You can also edit existing tests through conversation - describe changes, review a visual diff, and apply with one click. Application Analysis It can crawl and analyze your website to discover features, search through the knowledge graph to find specific functionality from previous crawls, and search your project’s knowledge base. This helps the assistant provide context-aware suggestions tailored to your application. Configuration Management The assistant can show you what configurations are available in your project, help you create new configurations for credentials and test data, and display interactive forms right in the chat to collect configuration details. It supports config types like username/password, username/password with 2FA, file uploads, API keys, and API calls. Issue Tracking If you have Linear or Jira connected, the assistant can search for issues by keyword, fetch a specific issue by key (e.g.PROJ-42), list recent issues, and create new ones in your configured team or project. For Jira, it reads title, description, status, assignee, priority, issue type, and labels from your configured project — see Jira integration for what is and is not read. Jira users can also filter by issue type such as Epic, Bug, Story, or Task. If both integrations are connected, mention which tracker you want (e.g. “search Jira”); otherwise the assistant defaults to Linear.
Help & Support
The assistant can search QA.tech’s documentation to answer questions about platform features and capabilities, and provide you with contact information for customer support when you need help beyond what the AI can provide.
Safe Experimentation
Tests generated in chat aren’t committed until you explicitly click “Add Selected Tests” - you can review, edit, or reject them without affecting your project. Files uploaded in chat are isolated to that conversation only, so you can experiment freely without impacting teammates. See How Knowledge Works for details on chat isolation and when to move knowledge to your project.
How the Assistant Adapts
The Chat Assistant automatically adjusts its approach based on what you’re trying to accomplish:
You don’t need to specify how - just describe what you want naturally and the assistant handles the details.
What the Assistant Can Access
The assistant has access to your project’s data and can perform actions on your behalf:Custom MCP servers
Custom MCP servers let the AI chat use tools from services you already work with. You connect a remote MCP server, choose which of its tools the chat agent may use, and the agent picks them up alongside its built-in tools. Ask it to sync test cases, look up records, or file items in your other systems, and it calls your server’s tools to do it.This is the reverse direction of the MCP Server
integration. That one lets your AI editor call QA.tech. This one lets the
QA.tech chat call your services.
Set it up
1
Add a server
Go to Settings → Integrations → Custom MCP Integrations and click Add
server. Give it a name and enter the server URL, which must be a
Streamable HTTP MCP endpoint reachable over HTTPS.
2
Choose authentication
OAuth is the default: after saving, you sign in to the service in your
browser and QA.tech handles the rest — no keys to paste. See OAuth
below for how it works. If your server uses static credentials instead, pick
Bearer token or API key and paste the credential your server expects. It
is sent as an
Authorization: Bearer header on every request and stored
encrypted. Servers without authentication are also supported, and you can
add extra headers such as X-Api-Key under Additional headers if your
server needs them.3
Test the connection
Click Test connection. QA.tech connects to the server, lists its tools,
and shows each one with a permission control.
4
Set tool permissions
Decide per tool whether the chat agent may use it freely, must ask you
first, or may not use it at all. New tools default to Needs approval.
5
Use it in chat
Open the project chat and ask for something the tools can do, for example
“look up the TestRail cases for the checkout suite”. Enabled tools are
available to the agent automatically.
OAuth
Many MCP servers do not issue static API keys and instead use OAuth — the same flow as a “Sign in with …” button. QA.tech implements the MCP authorization specification, so connecting is fully automatic:- No configuration. QA.tech discovers the server’s OAuth settings and registers itself as a client. You never enter endpoints or client IDs.
- Browser sign-in. Add and connect (or Connect on the server card) sends you to the service’s own sign-in page, and back to QA.tech when you approve. Your password never passes through QA.tech.
- Automatic refresh. Tokens are stored encrypted and refreshed in the background. You only sign in again if the service revokes access — the server card then shows OAuth not connected.
Who connects
OAuth servers ask who connects:- Each member connects their own account (the default). Tools run with each person’s own permissions in the connected service, so nobody acts through a teammate’s account. When a member opens a chat before connecting, a banner above the message box asks them to sign in — one click, the service’s sign-in page, and straight back to the chat. If a connection stops working mid-conversation (for example after a password change), the failed tool call shows a Reconnect button right in the chat.
- Everyone shares one connection. One account — typically a service account — is used for the whole project. Whoever clicks Connect provides it. Pick this for servers where individual identity doesn’t matter.
Tool permissions
Every tool runs with the permission you assign. The server-level default applies to tools that appear later, so a server update never silently gains unrestricted access.
An approval request expires after 4 minutes. If you deny it or let it expire,
the tool is not executed and the agent is told the call was not permitted.
Requirements and limits
If a server exposes more tools than the limit, the extra tools are skipped.
Security
- Credentials are stored encrypted and are never shown again after saving. Editing a server keeps the stored values unless you enter new ones.
- OAuth uses the standard authorization-code flow with PKCE. Tokens and the client registration are stored encrypted, and access tokens are refreshed automatically. Signing in happens on the service’s own pages, so QA.tech never sees your password.
- Server URLs must resolve to public addresses. Private and internal network ranges are blocked, and redirects are followed at most 5 times with the same checks on every hop.
- Tool output is treated as untrusted input: the agent is instructed not to follow instructions embedded in it.
- An unreachable or misconfigured server never breaks the chat. Its tools are simply unavailable for that conversation. Run Test connection to see the error.
Troubleshooting
For anything else, contact support.