Who this is for
Backend / integration developers
Use the Open API to create and update CRM contacts, submit forms, manage
contact files, and run AI document extraction from your own backend.
Front-end / web developers
Use the Widget JS API to embed the chat bubble and control it
programmatically with
A2V2Widget.init() / .destroy().The two surfaces
| Surface | Auth | What it’s for | Reference |
|---|---|---|---|
| Open API (REST) | API key + secret (header) | Server-to-server: contacts, forms, contact files, document extraction | Endpoints |
| Widget JS API | None (public, per agent) | Embed and control the chat widget in the browser | Widget JS API |
Base URL
All Open API endpoints live under the/v1/open path on the A2V2.ai API host:
https://api.a2v2.ai is the production API host. Every example in this section uses it
as the base URL, followed by the fixed /v1/open path./v1/open path:
Authentication at a glance
Open API requests authenticate with an API key and secret sent as request headers. You create keys in the dashboard under Settings → API Permissions, and each key is scoped to a single agent with explicit permissions.Response format
Every Open API response is JSON with asuccess flag.
- Success
- Error
pagination object alongside data:
Rate limits
Each API key has its own rate limits, set when you create the key. The defaults are 60 requests per minute and 1,000 requests per hour, and you can raise them when configuring the key (up to 1,000/minute and 10,000/hour). Requests over the limit return an error response — back off and retry.Health check
A lightweight, unauthenticated endpoint confirms the API is reachable:Related
Authentication
Create an API key and scope its permissions.
Endpoints
Every Open API endpoint, grouped by domain.
Widget JS API
Embed and control the chat widget in the browser.
Embed & Install
The no-code way to add your agent to a site.