API: Open Chat Studio v1
Description: Build, deploy and monitor chatbots.

TAG: OpenAI
Description: OpenAI-compatible endpoints for interacting with chatbots.

ENDPOINTS:

POST /api/openai/{experiment_id}/chat/completions
  Summary: Chat Completions API for Experiments
  Description: 
Use OpenAI's client to send messages to the experiment and get responses. This will
create a new session in the experiment with all the provided messages
and return the response from the experiment.

The last message must be a 'user' message.

Example (Python):

```python
experiment_id = "your experiment ID"

client = OpenAI(
    api_key="your API key",
    base_url=f"https://chatbots.dimagi.com/api/openai/{experiment_id}",
)

completion = client.chat.completions.create(
    model="anything",
    messages=[
        {"role": "assistant", "content": "How can I help you today?"},
        {"role": "user", "content": "I need help with something."},
    ],
)

reply = completion.choices[0].message
```

  Parameters:
    - experiment_id (path, string (required)): Experiment ID
  Request Body:
    Content: application/json
    Schema: CreateChatCompletionRequest
  Responses:
    200: 
      Content: application/json
      Schema: CreateChatCompletionResponse
  Security:
    OAuth2
      - Required scopes: chatbots:interact
    apiKeyAuth
    tokenAuth

POST /api/openai/{experiment_id}/v{version}/chat/completions
  Summary: Versioned Chat Completions API for Experiments
  Description: 
Use OpenAI's client to send messages to the experiment and get responses. This will
create a new session in the experiment with all the provided messages
and return the response from the experiment.

The last message must be a 'user' message.

Example (Python):

```python
experiment_id = "your experiment ID"

client = OpenAI(
    api_key="your API key",
    base_url=f"https://chatbots.dimagi.com/api/openai/{experiment_id}/v{version}",
)

completion = client.chat.completions.create(
    model="anything",
    messages=[
        {"role": "assistant", "content": "How can I help you today?"},
        {"role": "user", "content": "I need help with something."},
    ],
)

reply = completion.choices[0].message
```

  Parameters:
    - experiment_id (path, string (required)): Experiment ID
    - version (path, integer (required)): No description
    - version (query, string (optional)): Version of experiment
  Request Body:
    Content: application/json
    Schema: CreateChatCompletionRequest
  Responses:
    200: 
      Content: application/json
      Schema: CreateChatCompletionResponse
  Security:
    OAuth2
      - Required scopes: chatbots:interact
    apiKeyAuth
    tokenAuth

SCHEMAS:

ChatCompletionResponseChoices:
  - finish_reason: string (required)
  - index: integer (required)
  - message: ChatCompletionResponseMessage (required)

ChatCompletionResponseMessage:
  - role: ChatCompletionResponseMessageRoleEnum (required)
  - content: string (required)

ChatCompletionResponseMessageRoleEnum:
  - enum: [assistant]

CreateChatCompletionRequest:
  - messages: array of Message (required)

CreateChatCompletionResponse:
  - id: string (required)
  - choices: array of ChatCompletionResponseChoices (required)
  - created: integer (required)
  - model: string (required)
  - object: ObjectEnum (required)

File:
  - name: string (required)
  - content_type: string
  - size: integer (required)
  - content_url: string (required)

Message:
  - created_at: string (required)
  - role: MessageRoleEnum (required)
  - content: string (required)
  - metadata: Unknown (required)
  - tags: array of string (required)
  - attachments: array of File (required)

MessageRoleEnum:
  - enum: [system, user, assistant]

ObjectEnum:
  - enum: [chat.completion]

SECURITY:
- OAuth2
  - Authorization Code Flow (authorization url: /o/authorize/, token url: /o/token/)
- API Key authentication (header: X-api-key)
- API Key authentication (cookie: sessionid)
- API Key authentication (header: X-Embed-Key)
- HTTP bearer authentication