> For the complete documentation index, see [llms.txt](https://docs.elevateab.com/elevate-helpcenter/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api.md).

# External API: Read and Run Your Experiments

The Elevate A/B Testing External API gives your own tools access to your Shopify experiments: read results with any key, and create and run experiments with a write key.

The Elevate A/B Testing External API gives your own tools access to your Shopify store's experiments and their results, with the same numbers and statistics you see in Elevate A/B Testing. Use it to feed a data warehouse, build internal dashboards, or give an AI agent live experiment data.

Reading is a `GET` request that returns JSON. You authenticate with an API key that you generate in Elevate A/B Testing under **Settings**, then **API**. MCP connections are set up separately, on the **MCP** page in the sidebar: Claude and ChatGPT connect by logging in with Elevate A/B Testing, and other MCP clients use a key created for each connection. See [MCP Overview](https://docs.elevateab.com/elevate-helpcenter/mcp/mcp-overview.md).

To work with your experiments from your code editor, see [Connect Claude Code or Cursor](https://docs.elevateab.com/elevate-helpcenter/mcp/connect-claude-code-or-cursor.md).

The API key is read-only. To create, launch, pause, resume, and end experiments, create a separate write key. See [Write access](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api/write-access.md). No key can edit an experiment's setup or delete an experiment.

### Pages in this section

| Page                                                                                                                           | What it covers                                                     |
| ------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------ |
| [External API Quickstart](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api/overview.md)          | Create a key, find an experiment, and fetch its results            |
| [Write access](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api/write-access.md)                 | Create a write key, and create and run experiments safely          |
| [Authentication](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api/authentication.md)             | How API keys and write keys work, and how to generate them         |
| [Rate limits](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api/rate-limits.md)                   | Per-store limits, the cost of each endpoint, and how to retry      |
| [Errors](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api/errors.md)                             | The error format and every error code                              |
| [Dates and time zones](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api/dates-and-time-zones.md) | Date windows, time zones, and timeseries buckets                   |
| [Filters](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api/filters.md)                           | Narrow results by device, country, UTM, custom event, and more     |
| [Statistics](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api/statistics.md)                     | How to read the significance status, the winner, and probabilities |
| [API reference](https://docs.elevateab.com/elevate-helpcenter/developer-reference/external-api/api-reference.md)               | Every endpoint, plus the OpenAPI document                          |
