> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mayaresearch.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Building with an AI agent

> One file to paste into Claude, Cursor or Copilot.

Give your coding agent this:

```
https://www.mayaresearch.ai/llm.txt
```

The same API described here, written for a model to read: endpoint, auth, every
voice and language, streaming format, error codes and working samples, with no
prose it has to guess around.

## Why a separate file

These pages are paginated, styled and cross-linked — good for a person, noise
for a model that wants the whole contract at once. `llm.txt` is one flat file
with no navigation to follow and nothing to infer.

## What agents get wrong

Worth pasting alongside it — these are the mistakes we see most:

<CardGroup cols={2}>
  <Card title="Writing PCM to .wav" icon="file-audio">
    The response has no file header. Renaming the file does not add one.
  </Card>

  <Card title="Skipping v2: true" icon="plug">
    Required on the first WebSocket frame, and `start` goes **before** you read
    `metadata`.
  </Card>

  <Card title="A new connection per request" icon="clock">
    Costs \~300 ms in handshake every time. Reuse it.
  </Card>

  <Card title="Lower-casing the voice" icon="quote-left">
    `ananya` is not `Ananya`. Voice names are case-sensitive.
  </Card>
</CardGroup>