Every Livqeno SDK and the CLI call the same REST API. There is no private
surface: what an SDK does, you can do with curl and the same
credential.
The reference pages below are generated from the API's controllers, so they list what the server actually serves — all 110 versioned endpoints, with each one's credential, rate limit and validated parameters.
Where to start
| If you want to | Read |
|---|---|
| Understand the base path, error shape and pagination | Conventions |
| Mint a token for a call | RTC API |
| Create conversations, post messages server-side | Chat API |
| Run a live stream | Live Streaming API |
| Read connection history and errors | Observability API |
| Register a webhook endpoint | Webhooks API |
| Manage projects, teammates and keys | Projects API |
| See everything at once | All endpoints |
Interactive schemas
The API serves its own OpenAPI document. On a local stack:
http://localhost:4100/docs
That is the place to read full response schemas and try a request against your own data. These pages cover request parameters, credentials and semantics; the OpenAPI document covers response shapes.
Which credential
Four credentials exist and none can act for another.
| Credential | Sent as | Used by |
|---|---|---|
| Project API key | Authorization: Bearer rvk_<env>_id.secret | Your backend |
| RTC token | Query parameter on the signaling socket | A browser or device in a call |
| Chat token | Authorization: Bearer …, or a query parameter on the chat socket | A browser or device in a conversation |
| Dashboard session | Authorization: Bearer <jwt> | The dashboard and the CLI |
Full model in Authentication.
Next steps
- Conventions — read this before the endpoint pages.
- Errors — every code the API returns.
- Limits & quotas — the ceilings you will meet.