Inference Studio

https://docs.inferencesolutions.com/
Admin
Inference Studio is a platform for designing and deploying interactive voice applications, contact...
Tokens:13,941
Snippets:39
Trust Score:7.5
Update:1 year ago
Show doc for...
Tokens:
Raw
### Create a Messaging Application using cURL

Source: https://context7.com/context7/inferencesolutions/llms.txt

This example shows how to create an SMS/MMS messaging application using the Inference Studio API. It configures a short code, defines keyword-based routing, and specifies responses or actions for each keyword. Includes a default response for un Orecognized messages.

```bash
# Create messaging application
curl -X POST https://api.inferencesolutions.com/v1/tasks/messaging \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "SMS Support Bot",
    "short_code": "54321",
    "keywords": [
      {
        "keyword": "HELP",
        "response": "Reply with HOURS for business hours, LOCATION for address, or TALK to speak with support"
      },
      {
        "keyword": "HOURS",
        "response": "We are open Monday-Friday 9AM-5PM EST"
      },
      {
        "keyword": "LOCATION",
        "response": "Visit us at 123 Main Street, Suite 100, New York, NY 10001"
      },
      {
        "keyword": "TALK",
        "action": "create_ticket",
        "response": "A support representative will call you within 1 hour"
      }
    ],
    "default_response": "Reply HELP for options"
  }'

# Response
{
  "task_id": "msg_xyz789",
  "short_code": "54321",
  "status": "active",
  "created_at": "2025-10-08T12:00:00Z"
}
```

--------------------------------

### Example Webhook Payload for Call Completion

Source: https://context7.com/context7/inferencesolutions/llms.txt

This JSON object represents an example payload received when a 'call.completed' event occurs. It includes event details, a timestamp, and call data such as call ID, duration, status, and any associated variables. A signature is also provided for verifying the payload's origin.

```json
{
  "event": "call.completed",
  "timestamp": "2025-10-08T14:35:00Z",
  "data": {
    "call_id": "call_123abc",
    "task_id": "tsk_abc123",
    "from": "+15551234567",
    "to": "+18005551234",
    "duration": 245,
    "status": "completed",
    "variables": {
      "menu_selection": "2",
      "agent_id": "agent_42"
    },
    "recording_url": "https://storage.example.com/recordings/call_123abc.wav"
  },
  "signature": "sha256=abc123..."
}
```

--------------------------------

### Authenticate and Get API Token using cURL

Source: https://context7.com/context7/inferencesolutions/llms.txt

Demonstrates how to authenticate with the Inference Studio API to retrieve an access token. This token is then used to authorize subsequent API requests. Requires a username and password for authentication.

```bash
# Authenticate and retrieve access token
curl -X POST https://api.inferencesolutions.com/v1/auth/login \
  -H "Content-Type: application/json" \
  -d '{
    "username": "your-username",
    "password": "your-password"
  }'

# Response
{
  "token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9...",
  "expires_in": 3600,
  "user_id": "usr_12345"
}
```

--------------------------------

### Authorize API Requests with Bearer Token using cURL

Source: https://context7.com/context7/inferencesolutions/llms.txt

Shows how to use the obtained access token to make authenticated GET requests to the Inference Studio API. The token is included in the 'Authorization' header as a Bearer token. This is essential for accessing protected resources like project lists.

```bash
# Use token for authenticated requests
curl -X GET https://api.inferencesolutions.com/v1/projects \
  -H "Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9..." \
  -H "Content-Type: application/json"
```

--------------------------------

### Postman Collection for Inference Studio API

Source: https://context7.com/context7/inferencesolutions/llms.txt

This JSON structure defines a Postman collection for interacting with the Inference Studio API. It includes authentication settings using a bearer token, and an example request for the 'Login' endpoint. Environment variables for base URL and credentials are also defined.

```json
{
  "info": {
    "name": "Inference Studio API",
    "schema": "https://schema.getpostman.com/json/collection/v2.1.0/collection.json"
  },
  "auth": {
    "type": "bearer",
    "bearer": [
      {
        "key": "token",
        "value": "{{access_token}}",
        "type": "string"
      }
    ]
  },
  "item": [
    {
      "name": "Authentication",
      "item": [
        {
          "name": "Login",
          "request": {
            "method": "POST",
            "url": "{{base_url}}/v1/auth/login",
            "body": {
              "mode": "raw",
              "raw": "{\n  \"username\": \"{{username}}\",\n  \"password\": \"{{password}}\"\n}"
            }
          }
        }
      ]
    }
  ],
  "variable": [
    {"key": "base_url", "value": "https://api.inferencesolutions.com"},
    {"key": "access_token", "value": ""},
    {"key": "username", "value": "your-username"},
    {"key": "password", "value": "your-password"}
  ]
}
```

--------------------------------

### Get Specific Call Details using cURL

Source: https://context7.com/context7/inferencesolutions/llms.txt

Retrieves detailed information for a specific call using its unique call ID. This includes flow trace, recording URL, and transcript if available. Requires an API token.

```bash
curl -X GET https://api.inferencesolutions.com/v1/logs/calls/call_123abc \
  -H "Authorization: Bearer YOUR_TOKEN"
```

--------------------------------

### Configure Play Audio Node

Source: https://context7.com/context7/inferencesolutions/llms.txt

Configures a 'play_audio' node to play an audio file from a specified URL. Supports repeating the audio and allowing interruptions. No external dependencies beyond the audio source.

```javascript
// Play Audio Node
{
  "node_type": "play_audio",
  "config": {
    "audio_source": "url",
    "audio_url": "https://storage.example.com/prompts/welcome.wav",
    "repeat": 1,
    "allow_interrupt": true
  }
}
```

--------------------------------

### Postman Integration

Source: https://context7.com/context7/inferencesolutions/llms.txt

Guidance on integrating with Inference Solutions Studio APIs using Postman, including collection structure and pre-request scripts for authentication.

```APIDOC
### Postman Collection Structure

This JSON defines a Postman collection for interacting with the Inference Studio API. It includes authentication settings and a sample login request.

```json
{
  "info": {
    "name": "Inference Studio API",
    "schema": "https://schema.getpostman.com/json/collection/v2.1.0/collection.json"
  },
  "auth": {
    "type": "bearer",
    "bearer": [
      {
        "key": "token",
        "value": "{{access_token}}",
        "type": "string"
      }
    ]
  },
  "item": [
    {
      "name": "Authentication",
      "item": [
        {
          "name": "Login",
          "request": {
            "method": "POST",
            "url": "{{base_url}}/v1/auth/login",
            "body": {
              "mode": "raw",
              "raw": "{\n  \"username\": \"{{username}}\",\n  \"password\": \"{{password}}\"\n}"
            }
          }
        }
      ]
    }
  ],
  "variable": [
    {"key": "base_url", "value": "https://api.inferencesolutions.com"},
    {"key": "access_token", "value": ""},
    {"key": "username", "value": "your-username"},
    {"key": "password", "value": "your-password"}
  ]
}
```

### Postman Pre-request Script (Automatic Token Refresh)

This script automatically handles the refresh of access tokens before making API requests if the current token is expired or missing.

```javascript
if (!pm.environment.get('access_token') || pm.environment.get('token_expiry') < Date.now()) {
    pm.sendRequest({
        url: pm.environment.get('base_url') + '/v1/auth/login',
        method: 'POST',
        header: {'Content-Type': 'application/json'},
        body: {
            mode: 'raw',
            raw: JSON.stringify({
                username: pm.environment.get('username'),
                password: pm.environment.get('password')
            })
        }
    }, function (err, response) {
        const jsonResponse = response.json();
        pm.environment.set('access_token', jsonResponse.token);
        pm.environment.set('token_expiry', Date.now() + (jsonResponse.expires_in * 1000));
    });
}
```
```

--------------------------------

### Create or Update Datastore Entry using cURL

Source: https://context7.com/context7/inferencesolutions/llms.txt

Creates a new entry or updates an existing one in a specified datastore. Requires the datastore name, a unique key, and data in JSON format. An API token is needed for authentication.

```bash
curl -X PUT https://api.inferencesolutions.com/v1/datastores/customer_profiles/key/user_12345 \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "John Smith",
    "phone": "+15551234567",
    "preferred_language": "en",
    "account_status": "active",
    "last_contact": "2025-10-07T14:30:00Z"
  }'
```

--------------------------------

### Voice Task Creation

Source: https://context7.com/context7/inferencesolutions/llms.txt

Create and deploy interactive voice tasks, defining call flow logic, prompts, and settings for IVR systems.

```APIDOC
## Voice Task Creation

### Description
Create and deploy interactive voice tasks, defining call flow logic, prompts, and settings for IVR systems.

### Method
POST

### Endpoint
/v1/tasks/voice

### Parameters
#### Request Body
- **name** (string) - Required - The name of the voice task.
- **description** (string) - Optional - A description for the voice task.
- **entry_point** (string) - Required - The ID of the starting node for the call flow.
- **nodes** (array) - Required - An array of node objects defining the call flow logic.
  - **id** (string) - Required - Unique identifier for the node.
  - **type** (string) - Required - The type of node (e.g., `play_audio`, `gather_input`, `transfer`).
  - **audio_url** (string) - Optional - URL for audio playback (for `play_audio` type).
  - **prompt** (string) - Optional - User prompt for input (for `gather_input` type).
  - **timeout** (integer) - Optional - Input timeout in seconds (for `gather_input` type).
  - **routes** (object) - Optional - Mapping of user input to next nodes (for `gather_input` type).
  - **destination** (string) - Optional - Phone number or URI for transfer (for `transfer` type).
- **settings** (object) - Optional - Configuration settings for the voice task.
  - **record_calls** (boolean) - Optional - Whether to record calls.
  - **max_duration** (integer) - Optional - Maximum call duration in seconds.

### Request Example
```json
{
  "name": "Customer Support IVR",
  "description": "Main customer support auto attendant",
  "entry_point": "welcome_prompt",
  "nodes": [
    {
      "id": "welcome_prompt",
      "type": "play_audio",
      "audio_url": "https://storage.example.com/prompts/welcome.wav",
      "next": "main_menu"
    },
    {
      "id": "main_menu",
      "type": "gather_input",
      "prompt": "Press 1 for sales, 2 for support, 3 for billing",
      "timeout": 5,
      "routes": {
        "1": "sales_queue",
        "2": "support_queue",
        "3": "billing_queue"
      }
    },
    {
      "id": "sales_queue",
      "type": "transfer",
      "destination": "+15551234567"
    }
  ],
  "settings": {
    "record_calls": true,
    "max_duration": 1800
  }
}
```

### Response
#### Success Response (200)
- **task_id** (string) - The unique identifier for the created voice task.
- **status** (string) - The current status of the task (e.g., `active`).
- **phone_number** (string) - The assigned phone number for the task.
- **created_at** (string) - The timestamp when the task was created.

#### Response Example
```json
{
  "task_id": "tsk_abc123",
  "status": "active",
  "phone_number": "+18005551234",
  "created_at": "2025-10-08T12:00:00Z"
}
```
```

--------------------------------

### Configure Speech Recognition Node

Source: https://context7.com/context7/inferencesolutions/llms.txt

Configures a 'speech_input' node for voice command recognition. It includes prompts, language settings, optional hints for better recognition, timeout, confidence threshold, and the variable to store recognized speech. Supports specific language codes.

```javascript
// Speech Recognition Node
{
  "node_type": "speech_input",
  "config": {
    "prompt": "Please say the department you'd like to reach",
    "language": "en-US",
    "hints": ["sales", "support", "billing", "technical"],
    "timeout": 5,
    "confidence_threshold": 0.7,
    "variable_name": "department_spoken"
  }
}
```

--------------------------------

### Studio API Authentication

Source: https://context7.com/context7/inferencesolutions/llms.txt

Authenticate with the Studio API to obtain an access token for subsequent requests. The token is used for authorizing all platform resource access.

```APIDOC
## Studio API Authentication

### Description
Authenticate with the Studio API to obtain an access token for subsequent requests. The token is used for authorizing all platform resource access.

### Method
POST

### Endpoint
/v1/auth/login

### Parameters
#### Request Body
- **username** (string) - Required - The username for authentication.
- **password** (string) - Required - The password for authentication.

### Request Example
```json
{
  "username": "your-username",
  "password": "your-password"
}
```

### Response
#### Success Response (200)
- **token** (string) - The access token for API requests.
- **expires_in** (integer) - The token's expiration time in seconds.
- **user_id** (string) - The ID of the authenticated user.

#### Response Example
```json
{
  "token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9...",
  "expires_in": 3600,
  "user_id": "usr_12345"
}
```

### Usage Example (Authenticated Request)
```bash
curl -X GET https://api.inferencesolutions.com/v1/projects \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json"
```
```

--------------------------------

### Create a Voice Task using cURL

Source: https://context7.com/context7/inferencesolutions/llms.txt

This snippet illustrates how to create a new interactive voice task (IVR) via the Inference Studio API. It defines the task's name, description, entry point, and the sequence of nodes (prompts, input gathering, transfers) that constitute the call flow. Includes settings for call recording and maximum duration.

```bash
# Create a new voice task
curl -X POST https://api.inferencesolutions.com/v1/tasks/voice \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Customer Support IVR",
    "description": "Main customer support auto attendant",
    "entry_point": "welcome_prompt",
    "nodes": [
      {
        "id": "welcome_prompt",
        "type": "play_audio",
        "audio_url": "https://storage.example.com/prompts/welcome.wav",
        "next": "main_menu"
      },
      {
        "id": "main_menu",
        "type": "gather_input",
        "prompt": "Press 1 for sales, 2 for support, 3 for billing",
        "timeout": 5,
        "routes": {
          "1": "sales_queue",
          "2": "support_queue",
          "3": "billing_queue"
        }
      },
      {
        "id": "sales_queue",
        "type": "transfer",
        "destination": "+15551234567"
      }
    ],
    "settings": {
      "record_calls": true,
      "max_duration": 1800
    }
  }'

# Response
{
  "task_id": "tsk_abc123",
  "status": "active",
  "phone_number": "+18005551234",
  "created_at": "2025-10-08T12:00:00Z"
}
```

--------------------------------

### Retrieve Datastore Entry using cURL

Source: https://context7.com/context7/inferencesolutions/llms.txt

Fetches a specific entry from a datastore using its name and unique key. Requires an API token for authorization. Returns the stored data along with creation and update timestamps.

```bash
curl -X GET https://api.inferencesolutions.com/v1/datastores/customer_profiles/key/user_12345 \
  -H "Authorization: Bearer YOUR_TOKEN"
```

--------------------------------

### Create Open Form with API

Source: https://context7.com/context7/inferencesolutions/llms.txt

This snippet demonstrates how to create a new open form using a POST request to the Inference Solutions API. It requires an authorization token and a JSON payload defining the form's structure, including fields, submission webhook, and thank you message. The response includes the form ID and public URL.

```shell
curl -X POST https://api.inferencesolutions.com/v1/forms \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Customer Feedback Survey",
    "form_type": "survey",
    "fields": [
      {
        "name": "satisfaction_rating",
        "type": "scale",
        "label": "How satisfied were you with our service?",
        "min": 1,
        "max": 5,
        "required": true
      },
      {
        "name": "comments",
        "type": "text",
        "label": "Additional comments",
        "max_length": 500,
        "required": false
      },
      {
        "name": "contact_back",
        "type": "boolean",
        "label": "May we contact you about your feedback?",
        "required": true
      }
    ],
    "submission_webhook": "https://api.example.com/feedback",
    "thank_you_message": "Thank you for your feedback!"
  }'
```

--------------------------------

### Query Datastore Entries using cURL

Source: https://context7.com/context7/inferencesolutions/llms.txt

Queries entries within a datastore based on specified filter criteria and a limit for the number of results. Requires an API token and the query parameters in JSON format.

```bash
curl -X POST https://api.inferencesolutions.com/v1/datastores/customer_profiles/query \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "filter": {
      "account_status": "active",
      "preferred_language": "en"
    },
    "limit": 100
  }'
```

--------------------------------

### Messaging Application

Source: https://context7.com/context7/inferencesolutions/llms.txt

Build and deploy SMS/MMS messaging flows, defining keyword routing and response templates for automated messaging.

```APIDOC
## Messaging Application

### Description
Build and deploy SMS/MMS messaging flows, defining keyword routing and response templates for automated messaging.

### Method
POST

### Endpoint
/v1/tasks/messaging

### Parameters
#### Request Body
- **name** (string) - Required - The name of the messaging application.
- **short_code** (string) - Required - The short code for sending and receiving messages.
- **keywords** (array) - Optional - An array of keyword objects for routing incoming messages.
  - **keyword** (string) - Required - The keyword to match.
  - **response** (string) - Optional - The text response for the keyword.
  - **action** (string) - Optional - An action to perform when the keyword is matched (e.g., `create_ticket`).
- **default_response** (string) - Optional - The response sent when no keyword matches.

### Request Example
```json
{
  "name": "SMS Support Bot",
  "short_code": "54321",
  "keywords": [
    {
      "keyword": "HELP",
      "response": "Reply with HOURS for business hours, LOCATION for address, or TALK to speak with support"
    },
    {
      "keyword": "HOURS",
      "response": "We are open Monday-Friday 9AM-5PM EST"
    },
    {
      "keyword": "LOCATION",
      "response": "Visit us at 123 Main Street, Suite 100, New York, NY 10001"
    },
    {
      "keyword": "TALK",
      "action": "create_ticket",
      "response": "A support representative will call you within 1 hour"
    }
  ],
  "default_response": "Reply HELP for options"
}
```

### Response
#### Success Response (200)
- **task_id** (string) - The unique identifier for the created messaging task.
- **short_code** (string) - The short code associated with the task.
- **status** (string) - The current status of the task (e.g., `active`).
- **created_at** (string) - The timestamp when the task was created.

#### Response Example
```json
{
  "task_id": "msg_xyz789",
  "short_code": "54321",
  "status": "active",
  "created_at": "2025-10-08T12:00:00Z"
}
```
```

--------------------------------

### Natural Language Auto Attendant

Source: https://context7.com/context7/inferencesolutions/llms.txt

Build intelligent auto attendants using Cloud Speech-to-Text and tone analysis integration. This allows for more natural interaction with users by understanding their spoken intent and sentiment.

```APIDOC
## Natural Language Auto Attendant

Build intelligent auto attendants using Cloud Speech-to-Text and tone analysis integration.

### Workflow Configuration

Defines the structure and behavior of a Natural Language IVR.

#### Fields

*   **task_name** (string) - The name of the IVR task.
*   **entry_node** (string) - The ID of the first node to execute.
*   **nodes** (array of objects) - An array of node configurations.
    *   **id** (string) - Required - Unique identifier for the node.
    *   **type** (string) - Required - The type of the node (e.g., 'play_audio', 'speech_recognition', 'ai_intent_detection', 'condition').
    *   **audio_url** (string) - (For 'play_audio' type) URL of the audio file.
    *   **next** (string) - (For 'play_audio', 'speech_recognition' types) The ID of the next node.
    *   **config** (object) - Configuration specific to the node type.
        *   (For 'speech_recognition' type):
            *   `provider` (string) - Speech recognition provider (e.g., 'google_cloud').
            *   `language` (string) - Language code (e.g., 'en-US').
            *   `enable_tone_analysis` (boolean) - Whether to enable tone analysis.
            *   `timeout` (integer) - Timeout in seconds.
            *   `variable_name` (string) - Variable to store the recognized speech.
        *   (For 'ai_intent_detection' type):
            *   `input_variable` (string) - Variable containing the input speech.
            *   `intents` (object) - Mapping of intent names to keywords.
            *   `tone_variable` (string) - Variable to store the detected tone.
            *   `routes` (object) - Mapping of intents/tones to next node IDs.
        *   (For 'condition' type):
            *   `variable` (string) - Variable to evaluate.
            *   `conditions` (array of objects) - Conditions and their outcomes.
                *   `operator` (string) - Comparison operator.
                *   `value` (string) - Value to compare against.
                *   `next` (string) - Next node ID.

### Example Configuration

```json
{
  "task_name": "NL Auto Attendant",
  "entry_node": "greeting",
  "nodes": [
    {
      "id": "greeting",
      "type": "play_audio",
      "audio_url": "https://storage.example.com/greet.wav",
      "next": "listen_intent"
    },
    {
      "id": "listen_intent",
      "type": "speech_recognition",
      "config": {
        "provider": "google_cloud",
        "language": "en-US",
        "enable_tone_analysis": true,
        "timeout": 10,
        "variable_name": "user_speech"
      },
      "next": "analyze_intent"
    },
    {
      "id": "analyze_intent",
      "type": "ai_intent_detection",
      "config": {
        "input_variable": "user_speech",
        "intents": {
          "billing_inquiry": ["bill", "payment", "charge", "invoice"],
          "technical_support": ["broken", "not working", "error", "problem"],
          "sales": ["buy", "purchase", "order", "price"]
        },
        "tone_variable": "caller_sentiment",
        "routes": {
          "billing_inquiry": "billing_department",
          "technical_support": "tech_support",
          "sales": "sales_team",
          "unknown": "operator"
        }
      }
    },
    {
      "id": "billing_department",
      "type": "condition",
      "config": {
        "variable": "caller_sentiment",
        "conditions": [
          {"operator": "equals", "value": "frustrated", "next": "priority_billing"},
          {"operator": "default", "next": "standard_billing"}
        ]
      }
    }
  ]
}
```
```

--------------------------------

### Datastore Operations API

Source: https://context7.com/context7/inferencesolutions/llms.txt

Manage persistent data storage for application state, customer information, and custom variables.

```APIDOC
## PUT /v1/datastores/{datastore_name}/key/{key_name}

### Description
Creates or updates an entry in a specified datastore using a unique key.

### Method
PUT

### Endpoint
https://api.inferencesolutions.com/v1/datastores/{datastore_name}/key/{key_name}

### Parameters
#### Path Parameters
- **datastore_name** (string) - Required - The name of the datastore.
- **key_name** (string) - Required - The unique key for the entry.

#### Headers
- **Authorization** (string) - Required - Bearer token for authentication.
- **Content-Type** (string) - Required - Application JSON.

#### Request Body
- **(object)** - Required - The data to store. Schema depends on the datastore.

### Request Example
```json
{
  "name": "John Smith",
  "phone": "+15551234567",
  "preferred_language": "en",
  "account_status": "active",
  "last_contact": "2025-10-07T14:30:00Z"
}
```

### Response
#### Success Response (200)
- **datastore** (string) - The name of the datastore.
- **key** (string) - The key of the updated entry.
- **updated_at** (string) - Timestamp when the entry was updated (ISO 8601).

#### Response Example
```json
{
  "datastore": "customer_profiles",
  "key": "user_12345",
  "updated_at": "2025-10-08T12:00:00Z"
}
```
```

```APIDOC
## GET /v1/datastores/{datastore_name}/key/{key_name}

### Description
Retrieves an entry from a specified datastore using its unique key.

### Method
GET

### Endpoint
https://api.inferencesolutions.com/v1/datastores/{datastore_name}/key/{key_name}

### Parameters
#### Path Parameters
- **datastore_name** (string) - Required - The name of the datastore.
- **key_name** (string) - Required - The unique key of the entry to retrieve.

#### Headers
- **Authorization** (string) - Required - Bearer token for authentication.

### Response
#### Success Response (200)
- **key** (string) - The key of the retrieved entry.
- **data** (object) - The data stored for the entry.
- **created_at** (string) - Timestamp when the entry was created (ISO 8601).
- **updated_at** (string) - Timestamp when the entry was last updated (ISO 8601).

#### Response Example
```json
{
  "key": "user_12345",
  "data": {
    "name": "John Smith",
    "phone": "+15551234567",
    "preferred_language": "en",
    "account_status": "active",
    "last_contact": "2025-10-07T14:30:00Z"
  },
  "created_at": "2025-09-15T10:00:00Z",
  "updated_at": "2025-10-08T12:00:00Z"
}
```
```

```APIDOC
## POST /v1/datastores/{datastore_name}/query

### Description
Queries entries within a specified datastore based on filter criteria.

### Method
POST

### Endpoint
https://api.inferencesolutions.com/v1/datastores/{datastore_name}/query

### Parameters
#### Path Parameters
- **datastore_name** (string) - Required - The name of the datastore to query.

#### Headers
- **Authorization** (string) - Required - Bearer token for authentication.
- **Content-Type** (string) - Required - Application JSON.

#### Request Body
- **filter** (object) - Required - Criteria to filter the datastore entries.
- **limit** (integer) - Optional - Maximum number of results to return.

### Request Example
```json
{
  "filter": {
    "account_status": "active",
    "preferred_language": "en"
  },
  "limit": 100
}
```

### Response
#### Success Response (200)
- ****(array)** - A list of datastore entries matching the query criteria. The structure of each entry depends on the datastore's schema.

#### Response Example
```json
[
  {
    "key": "user_12345",
    "data": {
      "name": "John Smith",
      "phone": "+15551234567",
      "preferred_language": "en",
      "account_status": "active",
      "last_contact": "2025-10-07T14:30:00Z"
    },
    "created_at": "2025-09-15T10:00:00Z",
    "updated_at": "2025-10-08T12:00:00Z"
  }
]
```
```

--------------------------------

### Create Open Form

Source: https://context7.com/context7/inferencesolutions/llms.txt

This endpoint allows you to create new forms programmatically. You can define form fields, submission behavior, and a thank you message.

```APIDOC
## POST /v1/forms

### Description
Creates a new open form with specified fields and configurations.

### Method
POST

### Endpoint
https://api.inferencesolutions.com/v1/forms

### Parameters
#### Headers
- **Authorization** (string) - Required - Bearer token for authentication.
- **Content-Type** (string) - Required - application/json

#### Request Body
- **name** (string) - Required - The name of the form.
- **form_type** (string) - Required - The type of the form (e.g., "survey").
- **fields** (array) - Required - An array of field objects defining the form structure.
  - **name** (string) - Required - The internal name of the field.
  - **type** (string) - Required - The type of the field (e.g., "scale", "text", "boolean").
  - **label** (string) - Required - The display label for the field.
  - **min** (integer) - Optional - Minimum value for scale type fields.
  - **max** (integer) - Optional - Maximum value for scale type fields.
  - **max_length** (integer) - Optional - Maximum length for text type fields.
  - **required** (boolean) - Required - Whether the field is mandatory.
- **submission_webhook** (string) - Optional - URL to send form submission notifications.
- **thank_you_message** (string) - Optional - Message displayed after form submission.

### Request Example
```json
{
  "name": "Customer Feedback Survey",
  "form_type": "survey",
  "fields": [
    {
      "name": "satisfaction_rating",
      "type": "scale",
      "label": "How satisfied were you with our service?",
      "min": 1,
      "max": 5,
      "required": true
    },
    {
      "name": "comments",
      "type": "text",
      "label": "Additional comments",
      "max_length": 500,
      "required": false
    },
    {
      "name": "contact_back",
      "type": "boolean",
      "label": "May we contact you about your feedback?",
      "required": true
    }
  ],
  "submission_webhook": "https://api.example.com/feedback",
  "thank_you_message": "Thank you for your feedback!"
}
```

### Response
#### Success Response (200)
- **form_id** (string) - The unique identifier for the created form.
- **public_url** (string) - The public URL of the created form.
- **embed_code** (string) - The iframe embed code for the form.
- **created_at** (string) - Timestamp of form creation.

#### Response Example
```json
{
  "form_id": "frm_survey123",
  "public_url": "https://forms.inferencesolutions.com/frm_survey123",
  "embed_code": "<iframe src='https://forms.inferencesolutions.com/frm_survey123' width='100%' height='500'></iframe>",
  "created_at": "2025-10-08T12:00:00Z"
}
```
```

--------------------------------

### Configure Natural Language IVR Task

Source: https://context7.com/context7/inferencesolutions/llms.txt

Defines a 'Natural Language IVR' task with multiple nodes for handling user intent via speech. Includes audio playback, speech recognition with tone analysis, AI intent detection, and conditional routing based on sentiment. Uses Google Cloud for speech.

```javascript
// Natural Language IVR Configuration
{
  "task_name": "NL Auto Attendant",
  "entry_node": "greeting",
  "nodes": [
    {
      "id": "greeting",
      "type": "play_audio",
      "audio_url": "https://storage.example.com/greet.wav",
      "next": "listen_intent"
    },
    {
      "id": "listen_intent",
      "type": "speech_recognition",
      "config": {
        "provider": "google_cloud",
        "language": "en-US",
        "enable_tone_analysis": true,
        "timeout": 10,
        "variable_name": "user_speech"
      },
      "next": "analyze_intent"
    },
    {
      "id": "analyze_intent",
      "type": "ai_intent_detection",
      "config": {
        "input_variable": "user_speech",
        "intents": {
          "billing_inquiry": ["bill", "payment", "charge", "invoice"],
          "technical_support": ["broken", "not working", "error", "problem"],
          "sales": ["buy", "purchase", "order", "price"]
        },
        "tone_variable": "caller_sentiment",
        "routes": {
          "billing_inquiry": "billing_department",
          "technical_support": "tech_support",
          "sales": "sales_team",
          "unknown": "operator"
        }
      }
    },
    {
      "id": "billing_department",
      "type": "condition",
      "config": {
        "variable": "caller_sentiment",
        "conditions": [
          {"operator": "equals", "value": "frustrated", "next": "priority_billing"},
          {"operator": "default", "next": "standard_billing"}
        ]
      }
    }
  ]
}
```

--------------------------------

### Create Outbound Campaign using cURL

Source: https://context7.com/context7/inferencesolutions/llms.txt

Creates a new outbound campaign with specified contacts, scheduling, and settings. Requires an API token for authorization and sends campaign details in JSON format.

```bash
curl -X POST https://api.inferencesolutions.com/v1/campaigns \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Holiday Promotion Calls",
    "task_id": "tsk_abc123",
    "contacts": [
      {
        "phone": "+15551234567",
        "variables": {
          "first_name": "John",
          "account_id": "ACC123"
        }
      },
      {
        "phone": "+15559876543",
        "variables": {
          "first_name": "Jane",
          "account_id": "ACC456"
        }
      }
    ],
    "schedule": {
      "start_time": "2025-10-15T09:00:00Z",
      "end_time": "2025-10-15T17:00:00Z",
      "time_zone": "America/New_York"
    },
    "settings": {
      "max_concurrent_calls": 10,
      "retry_failed": true,
      "retry_attempts": 2
    }
  }'
```

--------------------------------

### Studio Node Configuration

Source: https://context7.com/context7/inferencesolutions/llms.txt

Configure individual node types within visual workflows for advanced call flow control. This section details various node types such as Play Audio, Gather Input, Speech Recognition, Condition, API Call, and Transfer.

```APIDOC
## Studio Node Configuration

Configure individual node types within visual workflows for advanced call flow control.

### Node Types

*   **play_audio**
    *   `audio_source` (string) - Source of the audio (e.g., 'url').
    *   `audio_url` (string) - URL of the audio file.
    *   `repeat` (integer) - Number of times to repeat the audio.
    *   `allow_interrupt` (boolean) - Whether the audio can be interrupted.

*   **gather_input**
    *   `prompt` (string) - The prompt to play to the user.
    *   `min_digits` (integer) - Minimum number of digits to collect.
    *   `max_digits` (integer) - Maximum number of digits to collect.
    *   `timeout` (integer) - Timeout in seconds for input.
    *   `finish_on_key` (string) - Key to finish input collection.
    *   `variable_name` (string) - Variable to store the collected input.

*   **speech_input**
    *   `prompt` (string) - The prompt to play to the user.
    *   `language` (string) - Language for speech recognition (e.g., 'en-US').
    *   `hints` (array of strings) - Hints for speech recognition.
    *   `timeout` (integer) - Timeout in seconds for speech input.
    *   `confidence_threshold` (float) - Minimum confidence score for recognition.
    *   `variable_name` (string) - Variable to store the recognized speech.

*   **condition**
    *   `variable` (string) - The variable to evaluate.
    *   `conditions` (array of objects) - List of conditions and their outcomes.
        *   `operator` (string) - The comparison operator (e.g., 'equals', 'default').
        *   `value` (string) - The value to compare against.
        *   `next` (string) - The next node ID if the condition is met.

*   **api_call**
    *   `method` (string) - HTTP method for the API call (e.g., 'POST').
    *   `url` (string) - The URL of the API endpoint.
    *   `headers` (object) - Headers for the API request.
    *   `body` (object) - Request body for the API call.
    *   `response_variable` (string) - Variable to store the API response.
    *   `timeout` (integer) - Timeout in seconds for the API call.

*   **transfer**
    *   `destination` (string) - The phone number or agent to transfer to.
    *   `type` (string) - Type of transfer (e.g., 'warm').
    *   `whisper_prompt` (string) - Prompt to play to the agent.
    *   `timeout` (integer) - Timeout in seconds for the transfer.
    *   `fallback_node` (string) - Node to transfer to if the primary transfer fails.
```

--------------------------------

### Create Outbound Campaign

Source: https://context7.com/context7/inferencesolutions/llms.txt

Creates a new outbound campaign for making calls to a list of contacts.

```APIDOC
## POST /v1/campaigns

### Description
Creates a new outbound campaign for making calls to a list of contacts, with options for scheduling and retry settings.

### Method
POST

### Endpoint
https://api.inferencesolutions.com/v1/campaigns

### Parameters
#### Headers
- **Authorization** (string) - Required - Bearer token for authentication.
- **Content-Type** (string) - Required - Application JSON.

#### Request Body
- **name** (string) - Required - The name of the campaign.
- **task_id** (string) - Required - The ID of the task associated with the campaign.
- **contacts** (array) - Required - A list of contacts to call.
  - **phone** (string) - Required - The phone number of the contact.
  - **variables** (object) - Optional - Key-value pairs for contact-specific variables.
- **schedule** (object) - Required - Campaign scheduling information.
  - **start_time** (string) - Required - ISO 8601 format start time.
  - **end_time** (string) - Required - ISO 8601 format end time.
  - **time_zone** (string) - Required - Time zone for the schedule.
- **settings** (object) - Optional - Campaign settings.
  - **max_concurrent_calls** (integer) - Optional - Maximum concurrent calls.
  - **retry_failed** (boolean) - Optional - Whether to retry failed calls.
  - **retry_attempts** (integer) - Optional - Number of retry attempts.

### Request Example
```json
{
  "name": "Holiday Promotion Calls",
  "task_id": "tsk_abc123",
  "contacts": [
    {
      "phone": "+15551234567",
      "variables": {
        "first_name": "John",
        "account_id": "ACC123"
      }
    },
    {
      "phone": "+15559876543",
      "variables": {
        "first_name": "Jane",
        "account_id": "ACC456"
      }
    }
  ],
  "schedule": {
    "start_time": "2025-10-15T09:00:00Z",
    "end_time": "2025-10-15T17:00:00Z",
    "time_zone": "America/New_York"
  },
  "settings": {
    "max_concurrent_calls": 10,
    "retry_failed": true,
    "retry_attempts": 2
  }
}
```

### Response
#### Success Response (200)
- **campaign_id** (string) - The ID of the created campaign.
- **status** (string) - The status of the campaign.
- **total_contacts** (integer) - The total number of contacts in the campaign.
- **estimated_duration** (integer) - The estimated duration of the campaign in minutes.
- **created_at** (string) - The timestamp when the campaign was created.

#### Response Example
```json
{
  "campaign_id": "cmp_def456",
  "status": "scheduled",
  "total_contacts": 2,
  "estimated_duration": 45,
  "created_at": "2025-10-08T12:00:00Z"
}
```
```

--------------------------------

### Create PCI-Compliant Payment Collection Node

Source: https://context7.com/context7/inferencesolutions/llms.txt

Uses curl to create a 'secure_payment' node for PCI-DSS compliant payment collection. Configures merchant details, payment gateway, fields to collect, prompts, amount variable, and success/failure nodes. Requires API token and endpoint.

```bash
# Configure payment collection node
curl -X POST https://api.inferencesolutions.com/v1/tasks/tsk_abc123/nodes \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "node_id": "payment_collection",
    "node_type": "secure_payment",
    "config": {
      "merchant_id": "merchant_12345",
      "payment_gateway": "stripe",
      "collect_fields": ["card_number", "expiry", "cvv"],
      "prompts": {
        "card_number": "Please enter your 16-digit card number",
        "expiry": "Enter the expiration date as 4 digits",
        "cvv": "Enter the 3-digit security code"
      },
      "amount_variable": "payment_amount",
      "success_node": "payment_success",
      "failure_node": "payment_retry"
    }
  }'
```
Tokens:
Raw