Nuance Labs Browser API

Called by your page or app with a session token (`st_…`).

GitHub OpenAPI

Operations 4

POST /conversation Create a conversation #
GET /conversation/{conversation_id} Get a conversation #
GET /conversation/{conversation_id}/events Follow a conversation #
POST /conversation/{conversation_id}/abandon Abandon a conversation #

Work with this as data

Every API here is available over the APIs.io API and to AI agents over MCP.

MCP server

One button, every client — Claude, Cursor, VS Code and the rest.

https://apis.io/mcp

Tools for apis

7 MCP tools reach this
  • find_apisBrowse and filter every API in the catalog.
  • get_api_artifactsOne API's artifacts, grouped by type.
  • get_openapiThe primary OpenAPI for this API.
  • find_similar_apisAPIs that look like this one.
  • apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.
  • resolveTurn a domain, URL or GitHub org into the provider it belongs to.
  • find_cohortsEvery scored population of providers in the catalog.
All 92 tools →

Call it yourself

curl for this page
This API
curl "https://apis.io/api/v1/apis/nuance-labs:nuance-labs-browser-api"
All apis
curl "https://apis.io/api/v1/apis?limit=25"

Calls without X-API-Key need no key. Calls with it need a paid plan — get a key.

Get an API key

Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. By signing up you are subscribing to the APIs.io and API Evangelist newsletters — you can unsubscribe from any issue. See our Privacy Policy and Terms.

A second provider on the same verified email joins the account you already have.

OpenAPI Specification

nuance-labs-browser-api-openapi.yml Raw ↑
openapi: 3.2.0
info:
  title: Nuance Labs Browser API
  version: '1'
  description: 'Operations tagged Browser across 2 of this provider''s published API definitions: nuance-labs-openapi.json, nuance-labs-openapi.yml. Each path carries the servers of the definition it was published in.'
servers:
- url: https://api.nuancelabs.ai
tags:
- name: Browser
  description: Called by your page or app with a session token (`st_…`).
paths:
  /conversation:
    post:
      operationId: create-a-conversation
      parameters: []
      requestBody:
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/CreateConversationRequest'
        required: true
      responses:
        '201':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ConversationResponse'
          description: Successful Response
      summary: Create a conversation
      tags:
      - Browser
      description: Creates a conversation with an unused session token, which spends it. The conversation runs with what the token was minted with.
      security:
      - sessionToken: []
    servers:
    - url: https://api.nuancelabs.ai
  /conversation/{conversation_id}:
    get:
      operationId: get-a-conversation
      parameters:
      - in: path
        name: conversation_id
        required: true
        schema:
          format: uuid
          type: string
        description: The conversation's id.
      responses:
        '200':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ConversationResponse'
          description: Successful Response
      summary: Get a conversation
      tags:
      - Browser
      description: Reads a conversation's current state.
      security:
      - sessionToken: []
    servers:
    - url: https://api.nuancelabs.ai
  /conversation/{conversation_id}/events:
    get:
      description: 'The conversation''s events after the id `after`, oldest first: that its AI persona is getting ready, the room to join, its time limit once the persona starts, and its end. Held open for up to 20 seconds while nothing happens; an empty list means ask again with the same `after`. An open request keeps the conversation alive, so keep one open from creation to the end.'
      operationId: follow-a-conversation
      parameters:
      - in: path
        name: conversation_id
        required: true
        schema:
          format: uuid
          type: string
        description: The conversation's id.
      responses:
        '200':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/SessionEvents'
          description: Successful Response
      summary: Follow a conversation
      tags:
      - Browser
      security:
      - sessionToken: []
    servers:
    - url: https://api.nuancelabs.ai
  /conversation/{conversation_id}/abandon:
    post:
      description: Gives up on a conversation that has not started, freeing its place under your limit on conversations at once. A conversation that started or ended is left as it is.
      operationId: abandon-a-conversation
      parameters:
      - in: path
        name: conversation_id
        required: true
        schema:
          format: uuid
          type: string
        description: The conversation's id.
      responses:
        '200':
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/QueueState'
          description: Successful Response
      summary: Abandon a conversation
      tags:
      - Browser
      security:
      - sessionToken: []
    servers:
    - url: https://api.nuancelabs.ai
components:
  schemas:
    EndedEvent:
      type: object
      description: The conversation is over. Nothing follows.
      properties:
        type:
          const: ended
          default: ended
          type: string
          description: '`ended`.'
        id:
          type: integer
          description: The event's id.
        reason:
          type: string
          description: 'Why: `time_limit`, `policy_violation`, `age_check`, `client_ended`, `client_silent` or another end.'
      required:
      - id
      - reason
      - type
    SessionEvents:
      type: object
      description: A conversation's events after the request's `after`.
      properties:
        events:
          items:
            discriminator:
              mapping:
                admitted: '#/components/schemas/AdmittedEvent'
                ended: '#/components/schemas/EndedEvent'
                preparing: '#/components/schemas/PreparingEvent'
                queued: '#/components/schemas/QueuedEvent'
                started: '#/components/schemas/StartedEvent'
              propertyName: type
            oneOf:
            - $ref: '#/components/schemas/QueuedEvent'
            - $ref: '#/components/schemas/PreparingEvent'
            - $ref: '#/components/schemas/AdmittedEvent'
            - $ref: '#/components/schemas/StartedEvent'
            - $ref: '#/components/schemas/EndedEvent'
          type: array
          description: Oldest first, each with an increasing `id`; empty when nothing happened while the request was held.
      required:
      - events
    StartedEvent:
      type: object
      description: The persona has started the conversation. Sent again if its time limit changes.
      properties:
        type:
          const: started
          default: started
          type: string
          description: '`started`.'
        id:
          type: integer
          description: 'The event''s id: pass the last one you saw as `after`.'
        seconds_remaining:
          anyOf:
          - type: integer
          - type: 'null'
          description: How long the conversation may run, from when this answer arrived; null when it has no time limit.
      required:
      - id
      - seconds_remaining
      - type
    ConversationResponse:
      type: object
      description: A conversation.
      properties:
        id:
          format: uuid
          type: string
          description: The conversation's id.
        status:
          $ref: '#/components/schemas/ConversationStatus'
          description: Where the conversation is in its life.
        created_at:
          format: date-time
          type: string
          description: When it was created (UTC).
        updated_at:
          format: date-time
          type: string
          description: When its status last changed (UTC).
      required:
      - created_at
      - id
      - status
      - updated_at
    AdmittedEvent:
      type: object
      description: 'The conversation''s room is ready: join it.'
      properties:
        type:
          const: admitted
          default: admitted
          type: string
          description: '`admitted`.'
        id:
          type: integer
          description: 'The event''s id: pass the last one you saw as `after`.'
        livekit_url:
          type: string
          description: The LiveKit server to join.
        token:
          type: string
          description: The participant token to join it with.
      required:
      - id
      - livekit_url
      - token
      - type
    CreateConversationRequest:
      type: object
      description: The conversation to create.
      properties:
        timezone:
          anyOf:
          - type: string
          - type: 'null'
          description: The person's IANA time zone, such as `Europe/London`, so the persona knows their local time.
        max_session_seconds:
          anyOf:
          - exclusiveMinimum: 0
            type: integer
          - type: 'null'
          description: The longest the conversation may run, in seconds. Held to the service's maximum, which is also what it runs for when you leave this out. The `started` event's `seconds_remaining` says the limit it runs with.
        green_screen:
          default: false
          type: boolean
          description: The persona over solid green (`#00FF00`), so your page can key it out and draw it over its own background. Only where your deployment offers it; some personas can't be drawn over green, and a conversation that asks for it on one of them is refused.
    ConversationStatus:
      enum:
      - created
      - ready
      - queued
      - admitted
      - connecting
      - connected
      - ended
      - failed
      - expired
      title: ConversationStatus
      type: string
      description: '`created`, `ready` and `queued` before a runtime is held for it, `admitted` once one is, `connecting` and `connected` as the browser or app joins, then `ended`. `failed` and `expired` mean it ended without running.'
    QueuedEvent:
      type: object
      description: 'Waiting for a slot. A conversation started with an API key or a session token never waits in a queue, so it never gets this event: it may get `preparing`, then gets `admitted`.'
      properties:
        type:
          const: queued
          default: queued
          type: string
          description: '`queued`.'
        id:
          type: integer
          description: 'The event''s id: pass the last one you saw as `after`.'
        position:
          anyOf:
          - type: integer
          - type: 'null'
          description: Its place in the queue, where 0 is next.
        estimated_wait_seconds:
          anyOf:
          - type: integer
          - type: 'null'
          description: Roughly how long until it is admitted, when known.
      required:
      - estimated_wait_seconds
      - id
      - position
      - type
    PreparingEvent:
      type: object
      description: A runtime is held for the conversation and is getting ready for it, which can take up to about half a minute; `admitted` follows. Sent once.
      properties:
        type:
          const: preparing
          default: preparing
          type: string
          description: '`preparing`.'
        id:
          type: integer
          description: 'The event''s id: pass the last one you saw as `after`.'
      required:
      - id
      - type
    QueueState:
      type: object
      description: Where a conversation stands.
      properties:
        status:
          $ref: '#/components/schemas/ConversationStatus'
          description: Where the conversation is in its life.
        seconds_remaining:
          anyOf:
          - type: integer
          - type: 'null'
          description: Time left in a connected conversation, when it has a time limit.
      required:
      - seconds_remaining
      - status
  securitySchemes:
    apiKey:
      type: http
      scheme: bearer
      description: An API key from your organization's API keys page on the platform, `nk_…`. Server only.
    sessionToken:
      type: http
      scheme: bearer
      description: A session token your server minted, `st_…`.
x-refined-from:
- nuance-labs-openapi.json
- nuance-labs-openapi.yml
x-provenance:
  method: harvested
  publisher: Nuance Labs
  source: https://docs.nuancelabs.ai/openapi.json
  note: split by refine-openapis from the provider's own published document, verified by fetch (openapi/_original/*.fetch.json);
    stamped by stamp-openapi-provenance.py (roadmap#793)