> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.qa.army/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.qa.army/_mcp/server.

# Choose a QA.army Interface

> Choose the app, CLI, or public API for setup, saved Web Tests, Runs, and evidence review.

Use the app for visual work, the CLI for terminal workflows, and the public API
when you need to integrate supported operations. They use the same model:

```text
Workspace -> Project -> Test Group -> Test -> Run
```

| Interface | Use it for                                                                | Start here                                  |
| --------- | ------------------------------------------------------------------------- | ------------------------------------------- |
| App       | Set up a Project, edit Tests, start Web Runs, and review evidence         | [Quick Start](/get-started/quick-start)     |
| CLI       | Save Tests and operate supported Web Runs from a terminal or coding agent | [Install and use the CLI](/interfaces/cli)  |
| API       | Integrate documented operations with an authorized credential             | [API Access](/interfaces/api-access)        |
| MCP       | Structured agent tools when a supported connection is available           | [MCP availability](/interfaces/mcp)         |
| Chatbot   | Conversational product access; currently unavailable                      | [Chatbot availability](/interfaces/chatbot) |

The [public website](https://www.qa.army/) explains the product. It does not
operate Workspace data.

## Choose access for the task

An agent's setup access covers discovery, saving Tests, and supported Run
operations. It does not grant all the permissions of a signed-in user. Updating
Tests, organizing Test Groups, or cancelling Runs requires a supported user
session or profile API key. Test Account access requires a signed-in user session.

Use the app when a task needs interactive authorization or evidence review.
Use the [coding-agent guide](/interfaces/coding-agents) for direct Test authoring,
safe retries, and reporting actual results. Changing interfaces does not expand
Workspace access or change a Run's verdict.

## Check target availability

Direct Web testing is available. Native Android, iOS, and Desktop execution are
not generally available. A mobile-sized browser viewport is a Web Test, not
proof of native app coverage. Local-only targets need a supported connection;
see [Local Development](/interfaces/local-development).