===============
LIBRARY RULES
===============
From library maintainers:
- Base URL is https://api.callmissed.com
- Endpoints are OpenAI-compatible: use the OpenAI SDK and set base_url to https://api.callmissed.com/v1
- API keys are prefixed with cm_ and sent as the header Authorization: Bearer cm_your_key
- Use sarvam-30b or sarvam-105b for Indic-language LLM tasks
- Use saaras:v3 for Indian-language speech-to-text across 22 Indic languages
- Use bulbul:v3 for Indian-language text-to-speech (39 voices across 11 languages)
- The Anthropic Messages API is available at /v1/messages using the Anthropic SDK
- Free-tier models include kimi-k2.5, sarvam-30b, sarvam-105b, saaras:v3, and bulbul:v3

### Install livekit-client

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-sessions-api.md

Install the necessary LiveKit client library for Node.js.

```bash
npm install livekit-client
```

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

### Install OpenAI SDK for Go

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Install the official OpenAI SDK for Go using go modules. This is the recommended SDK for most use cases.

```bash
go get github.com/openai/openai-go
```

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

### Install OpenAI SDK for PHP

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Install the official OpenAI SDK for PHP using Composer. This is the recommended SDK for most use cases.

```bash
composer require openai-php/client
```

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

### Install OpenAI SDK for Ruby

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Install the official OpenAI SDK for Ruby using RubyGems. This is the recommended SDK for most use cases.

```bash
gem install ruby-openai
```

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

### Install callmissed-docs-mcp Globally

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/mcp-server.md

Install the server globally if you prefer to have it available system-wide. Otherwise, npx can run it on demand without installation.

```bash
npm install -g callmissed-docs-mcp
```

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

### Install JavaScript SDK

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-sdk.md

Install the CallMissed JavaScript SDK using npm.

```bash
npm install @callmissed/voice
```

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

### Install Python SDK

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-sdk.md

Install the CallMissed Python SDK using pip.

```bash
pip install callmissed
```

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

### Install OpenAI SDK for Python

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Install the official OpenAI SDK for Python using pip. This is the recommended SDK for most use cases.

```bash
pip install openai
```

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

### Install OpenAI SDK for JavaScript/TypeScript

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Install the official OpenAI SDK for JavaScript and TypeScript using npm. This is the recommended SDK for most use cases.

```bash
npm install openai
```

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

### JavaScript SDK Quick Start

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-sdk.md

Initialize a VoiceSession, connect to the WebSocket, and handle events like transcripts and agent text. Disconnect when finished.

```typescript
import { VoiceSession } from "@callmissed/voice";

// Get wsUrl and token from your backend (POST /v1/voice/sessions)
const session = new VoiceSession({ wsUrl, token });

session.on("transcript", (text) => console.log("User:", text));
session.on("agentText", (text) => console.log("Agent:", text));
session.on("stateChange", (state) => updateUI(state));
session.on("error", (msg) => console.error(msg));

await session.connect(); // Starts mic + WebSocket

// Later...
session.disconnect();
```

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

### Install CallMissed SDK (Python)

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/quickstart.md

Install the CallMissed SDK for Python using pip. Ensure you have Python and pip installed.

```bash
pip install openai
```

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

### Full Python Example for Function Calling

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/chat-function-calling.md

This Python example demonstrates the complete workflow of function calling: sending an initial request with tools, checking for tool calls in the response, executing the function, and sending the result back to the model.

```python
# Step 1: Send initial request with tools
response = client.chat.completions.create(
    model="sarvam-30b",
    messages=[{"role": "user", "content": "What's the weather in Mumbai?"}],
    tools=[
        {
            "type": "function",
            "function": {
                "name": "get_weather",
                "description": "Get weather for a city",
                "parameters": {
                    "type": "object",
                    "properties": {"city": {"type": "string"}},
                    "required": ["city"]
                }
            }
        }
    ],
    tool_choice="auto"
)

# Step 2: Check if model wants to call a tool
msg = response.choices[0].message
if msg.tool_calls:
    tool_call = msg.tool_calls[0]
    # Execute your function here...
    result = get_weather(json.loads(tool_call.function.arguments)["city"])

    # Step 3: Send result back
    final = client.chat.completions.create(
        model="sarvam-30b",
        messages=[
            {"role": "user", "content": "What's the weather in Mumbai?"},
            msg,  # assistant message with tool_calls
            {"role": "tool", "tool_call_id": tool_call.id, "content": json.dumps(result)}
        ]
    )
    print(final.choices[0].message.content)
```

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

### Python SDK Quick Start

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-sdk.md

Connect to CallMissed, create a voice session, and stream audio. Handles user transcripts, agent responses, and audio playback.

```python
import asyncio
from callmissed import CallMissed, SessionConfig

async def main():
    async with CallMissed(jwt_token="your_token") as client:
        # Create session
        session = await client.voice.create_session(
            SessionConfig(
                system_prompt="You are a helpful assistant.",
                voice="shubh",
                llm_model="sarvam-30b",
            )
        )

        # Connect and stream
        async with client.voice.connect(session) as ws:
            ws.on("transcript", lambda t: print(f"User: {t}"))
            ws.on("agent_text", lambda t: print(t, end=""))
            ws.on("audio", lambda chunk: play_audio(chunk))
            ws.on("error", lambda msg: print(f"Error: {msg}"))

            await ws.send_audio(pcm_bytes)
            await asyncio.sleep(60)

        # Get transcript
        turns = await client.voice.get_transcript(session.id)
        for turn in turns:
            print(f"User: {turn.user_transcript}")
            print(f"Agent: {turn.agent_response}")

asyncio.run(main())
```

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

### Configure and Use OpenAI SDK in Go

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Configure the OpenAI client with your API key and base URL, then make a chat completion request. Ensure you have the 'openai-go' package installed.

```go
package main

import (
    "context"
    "fmt"
    "github.com/openai/openai-go"
    "github.com/openai/openai-go/option"
)

func main() {
    client := openai.NewClient(
        option.WithAPIKey("cm_your_api_key"),
        option.WithBaseURL("https://api.callmissed.com/v1"),
    )

    resp, _ := client.Chat.Completions.New(context.Background(),
        openai.ChatCompletionNewParams{
            Model: openai.F("sarvam-30b"),
            Messages: openai.F([]openai.ChatCompletionMessageParamUnion{
                openai.UserMessage("Hello"),
            }),
        },
    )
    fmt.Println(resp.Choices[0].Message.Content)
}
```

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

### Install CallMissed SDK (JavaScript/TypeScript)

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/quickstart.md

Install the CallMissed SDK for JavaScript or TypeScript using npm. This is required for Node.js environments.

```bash
npm install openai
```

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

### Install Anthropic SDK for JavaScript/TypeScript

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Install the Anthropic SDK for JavaScript/TypeScript using npm. This SDK is used for the /v1/messages endpoint.

```bash
npm install @anthropic-ai/sdk
```

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

### Install Anthropic SDK for Python

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Install the Anthropic SDK for Python using pip. This SDK is used for the /v1/messages endpoint.

```bash
pip install anthropic
```

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

### Install CallMissed SDK (PHP)

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/quickstart.md

Install the CallMissed SDK for PHP using Composer. This command adds the client to your project's dependencies.

```bash
composer require openai-php/client
```

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

### Install CallMissed SDK (Ruby)

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/quickstart.md

Install the CallMissed SDK for Ruby using the gem command. This makes the library available in your Ruby project.

```bash
gem install ruby-openai
```

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

### Upgrade OpenAI SDK for Go

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Upgrade the OpenAI SDK for Go to the latest version using go get.

```bash
go get -u github.com/openai/openai-go
```

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

### Install CallMissed SDK (Go)

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/quickstart.md

Add the CallMissed SDK for Go to your project dependencies. This command fetches the latest version.

```bash
go get github.com/openai/openai-go
```

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

### Configure and Use OpenAI SDK in Python

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Configure the OpenAI client with your API key and base URL, then make a chat completion request. Ensure you have the 'openai' package installed.

```python
from openai import OpenAI

client = OpenAI(
    api_key="cm_your_api_key",
    base_url="https://api.callmissed.com/v1"
)

response = client.chat.completions.create(
    model="sarvam-30b",
    messages=[{"role": "user", "content": "Hello"}]
)
print(response.choices[0].message.content)
```

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

### Example Request with Gateway Plugins

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/gateway-plugins.md

This JSON object demonstrates how to enable various gateway plugins like web search, file parsing, response healing, and context compression within a chat completion request.

```json
{
  "model": "anthropic/claude-sonnet-4.6",
  "messages": [{ "role": "user", "content": "Summarize this PDF" }],
  "plugins": [
    { "id": "web" },
    "file-parser",
    "response-healing",
    "context-compression"
  ]
}
```

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

### Configure and Use OpenAI SDK in PHP

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Configure the OpenAI client with your API key and base URL, then make a chat completion request. Ensure you have the 'openai-php/client' package installed.

```php
<?php
$client = OpenAI::factory()
    ->withApiKey('cm_your_api_key')
    ->withBaseUri('https://api.callmissed.com/v1')
    ->make();

$response = $client->chat()->create([
    'model'    => 'sarvam-30b',
    'messages' => [['role' => 'user', 'content' => 'Hello']],
]);

echo $response->choices[0]->message->content;
```

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

### Configure and Use OpenAI SDK in Ruby

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Configure the OpenAI client with your API key and base URL, then make a chat completion request. Ensure you have the 'ruby-openai' gem installed.

```ruby
require "openai"

client = OpenAI::Client.new(
  access_token: "cm_your_api_key",
  uri_base: "https://api.callmissed.com/v1"
)

response = client.chat(
  parameters: {
    model: "sarvam-30b",
    messages: [{ role: "user", content: "Hello" }]
  }
)
puts response.dig("choices", 0, "message", "content")
```

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

### Configure and Use OpenAI SDK in TypeScript

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/sdks.md

Configure the OpenAI client with your API key and base URL, then make a chat completion request. Ensure you have the 'openai' package installed.

```typescript
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "cm_your_api_key",
  baseURL: "https://api.callmissed.com/v1",
});

const response = await client.chat.completions.create({
  model: "sarvam-30b",
  messages: [{ role: "user", content: "Hello" }],
});
console.log(response.choices[0].message.content);
```

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

### Use kimi-k2.5 during maintenance

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/models-kimi-fast.md

While kimi-k2.5-fast is under maintenance, point your code at kimi-k2.5. This example demonstrates how to make a streaming chat completion request using the OpenAI Python client.

```python
from openai import OpenAI

client = OpenAI(
    base_url="https://api.callmissed.com/v1",
    api_key="cm_your_api_key",
)

response = client.chat.completions.create(
    model="kimi-k2.5",  # kimi-k2.5-fast is under maintenance
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain quantum computing briefly."},
    ],
    stream=True,
)

for chunk in response:
    print(chunk.choices[0].delta.content or "", end="", flush=True)
```

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

### Basic Chat Completion in JavaScript

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/chat-completion.md

This JavaScript example demonstrates how to interact with the Chat Completions API. Configure the `apiKey` and `baseURL` in the OpenAI client constructor. The response contains the AI's message content.

```javascript
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "cm_your_key",
  baseURL: "https://api.callmissed.com/v1",
});

const response = await client.chat.completions.create({
  model: "sarvam-30b",
  messages: [
    { role: "system", content: "You are a helpful assistant." },
    { role: "user", content: "What is the capital of India?" },
  ],
});

console.log(response.choices[0].message.content);
```

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

### Generate Image with cURL

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/image-generation.md

Make a POST request to the image generation endpoint using cURL. This example demonstrates how to send the prompt, model, and size parameters in the request body.

```Bash
curl -X POST https://api.callmissed.com/v1/images/generations \
  -H "Authorization: Bearer cm_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "flux-2-klein-9b",
    "prompt": "A golden retriever in a sunlit library",
    "n": 1,
    "size": "1024x1024"
  }'
```

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

### List Models and Context Windows

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/chat-completion.md

Demonstrates how to list available models and retrieve their context window sizes using the Python client.

```APIDOC
## List Models and Context Windows

### Description
This section describes how to retrieve model information, specifically the `context_window` (token count for prompt + completion), using the `GET /v1/models` endpoint. The response includes `context_window` and `context_length` for compatibility.

### Endpoint
`GET /v1/models`

### Response
#### Success Response (200)
- **models** (array) - A list of available models.
  - Each model object contains:
    - **id** (string) - The model identifier.
    - **context_window** (integer) - The canonical context window size in tokens.
    - **context_length** (integer) - An alias for `context_window` for OpenAI SDK compatibility.
    - **model_extra** (object) - Additional model-specific information.
      - **context_window** (integer) - The context window size.
      - **supports_vision** (boolean) - Indicates if the model supports vision.

### Request Example
```python
from openai import OpenAI

client = OpenAI(api_key="cm_your_key", base_url="https://api.callmissed.com/v1")

for m in client.models.list():
    extra = m.model_extra or {}
    print(m.id, extra.get("context_window"), extra.get("supports_vision"))
```
```

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

### Connect to LiveKit Room with livekit-client SDK

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-agent.md

Connect your client to the LiveKit room using the provided URL and token. This code sets up event listeners for subscribed tracks and received transcriptions, then enables the microphone. The agent joins automatically upon connection.

```javascript
import { Room, RoomEvent, Track } from "livekit-client";

const room = new Room();

room.on(RoomEvent.TrackSubscribed, (track, pub, participant) => {
  if (track.kind === Track.Kind.Audio) {
    const el = track.attach();
    document.body.appendChild(el);
  }
});

room.on(RoomEvent.TranscriptionReceived, (segments, participant) => {
  for (const seg of segments) {
    if (seg.final) {
      const who = participant?.isLocal ? "You" : "Agent";
      console.log(who + ": " + seg.text);
    }
  }
});

await room.connect(session.ws_url, session.token);
await room.localParticipant.setMicrophoneEnabled(true);
```

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

### List All Models

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/models.md

Use this endpoint to retrieve a list of all available models. No authentication is required.

```bash
# List all models
curl https://api.callmissed.com/api/v1/models
```

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

### Get Session

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-sessions-api.md

Retrieves details for a specific voice session by its ID.

```APIDOC
## Get Session

### Description
Retrieves detailed information about a specific voice session using its unique identifier.

### Method
GET

### Endpoint
/v1/voice/sessions/{id}

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier of the voice session.

### Request Example
```bash
curl https://api.callmissed.com/v1/voice/sessions/{id} \
  -H "Authorization: Bearer cm_your_api_key"
```

### Response
#### Success Response (200)
Returns a `VoiceSessionOut` object, excluding `ws_url` and `token`.
```

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

### Get Transcript

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-sessions-api.md

Retrieves the transcript for a specific voice session in various formats.

```APIDOC
## Get Transcript

### Description
Retrieves the transcript of a voice session. The transcript can be returned in JSON, plain text, or SRT format.

### Method
GET

### Endpoint
/v1/voice/sessions/{id}/transcript

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier of the voice session.

#### Query Parameters
- **format** (string) - Optional - The desired format for the transcript. Defaults to `json`. Allowed values: `json`, `txt`, `srt`.

### Request Example
```bash
curl "https://api.callmissed.com/v1/voice/sessions/{id}/transcript?format=json" \
  -H "Authorization: Bearer cm_your_api_key"
```

### Response
#### Success Response (200)
Returns the transcript in the specified format (`json`, `txt`, or `srt`). The JSON format returns an array of turns, each containing details like user transcript, agent response, and timing information.
```

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

### Get conversation messages

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/conversations.md

Retrieves the ordered list of messages for a specific conversation.

```APIDOC
## GET /api/v1/conversations/:id/messages

### Description
Retrieves the ordered list of messages for a specific conversation.

### Method
GET

### Endpoint
/api/v1/conversations/:id/messages

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier of the conversation.

### Response
#### Success Response (200)
- **id** (string) - The unique identifier for the message.
- **conversation_id** (string) - The identifier of the conversation this message belongs to.
- **role** (string) - The role of the sender (e.g., 'user', 'assistant').
- **content** (string) - The content of the message.
- **message_type** (string) - The type of the message (e.g., 'text').
- **tokens_used** (integer) - The number of tokens used for this message (null if not applicable).
- **created_at** (string) - The timestamp when the message was created.

#### Response Example
```json
[
  { "id": "...", "conversation_id": "c0ffee00-...", "role": "user", "content": "Where is my order?", "message_type": "text", "tokens_used": null, "created_at": "2026-06-06T11:55:00Z" },
  { "id": "...", "conversation_id": "c0ffee00-...", "role": "assistant", "content": "Let me check that for you.", "message_type": "text", "tokens_used": 18, "created_at": "2026-06-06T11:55:02Z" }
]
```
```

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

### Token Count Response Example

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/anthropic-api.md

The response from the count_tokens endpoint provides the estimated number of input tokens.

```json
{"input_tokens": 15}
```

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

### Get Specific Model Details

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/models.md

Fetch details for a particular model by providing its unique ID in the URL.

```bash
# Get a specific model
curl https://api.callmissed.com/api/v1/models/sarvam-30b
```

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

### Create Order for Credit Top-up

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/payments.md

Initiates a credit top-up order. Use the returned payment_session_id with the Cashfree SDK for checkout. Authentication is via JWT.

```bash
curl https://api.callmissed.com/api/v1/payments/create-order \
  -H "Authorization: Bearer <jwt_access_token>" \
  -H "Content-Type: application/json" \
  -d '{"type": "credit_topup", "credits": 5000}'
```

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

### Access Model Information with JavaScript

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/model-access.md

This JavaScript example shows how to fetch and process model access data using the Fetch API. It includes logic to list free LLM models and determine if a given model is accessible on the free plan.

```typescript
const resp = await fetch("https://api.callmissed.com/api/v1/models/access");
const data = await resp.json();

// List free LLM models
const freeLLMs = data.plans.free.by_category.llm;
console.log("Free LLM models:", freeLLMs);

// Check if a model is available on free plan
const model = "openai/gpt-5.4-mini";
const isFree = data.plans.free.models.includes(model);
console.log(model, "on free plan:", isFree); // false
```

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

### Get Model Access Information

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/model-access.md

Retrieves information about which models are accessible by each plan tier. This endpoint does not require authentication.

```APIDOC
## GET /api/v1/models/access

### Description
Retrieves a list of models accessible by each plan tier (Free, Starter, Pro, Enterprise), categorized by LLM, STT, TTS, and Image.

### Method
GET

### Endpoint
/api/v1/models/access

### Parameters
None

### Request Example
None

### Response
#### Success Response (200)
- **plans** (object) - An object containing details for each plan tier.
  - **free** (object) - Details for the free plan.
    - **models** (array) - List of all model IDs available on the free plan.
    - **by_category** (object) - Models categorized by type (llm, stt, tts, image).
    - **restriction** (string) - A summary of the plan's model restrictions.
  - **starter** (object) - Details for the starter plan (similar structure to free).
  - **pro** (object) - Details for the pro plan (similar structure to free).
  - **enterprise** (object) - Details for the enterprise plan (similar structure to free, may include custom models).

#### Response Example
```json
{
  "plans": {
    "free": {
      "models": ["auto", "sarvam-30b", ...],
      "by_category": {
        "llm": ["auto", "sarvam-30b", ...],
        "stt": ["saaras:v3", ...],
        "tts": ["bulbul:v3", ...],
        "image": ["flux-2-klein-9b", ...]
      },
      "restriction": "25 models across 4 categories"
    },
    "starter": {
      "models": ["...all free models + 300+ paid models"],
      "by_category": { "llm": ["..."], "stt": ["..."], "tts": ["..."], "image": ["..."] },
      "restriction": "All 300+ models"
    },
    "pro": {
      "models": ["..."],
      "by_category": { "llm": ["..."], "stt": ["..."], "tts": ["..."], "image": ["..."] },
      "restriction": "All 300+ models"
    },
    "enterprise": {
      "models": ["..."],
      "by_category": { "llm": ["..."], "stt": ["..."], "tts": ["..."], "image": ["..."] },
      "restriction": "All 300+ models + custom models"
    }
  }
}
```

### Error Handling
- **403 Forbidden**: Returned when a free-plan user attempts to access a paid model. The response body will contain an error message specifying the model and a link to upgrade.
```

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

### Get Specific Model Details

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/models.md

Retrieves detailed information about a single, specific AI model using its unique ID.

```APIDOC
## GET /api/v1/models/{model_id}

### Description
Retrieves detailed information for a specific model identified by its ID.

### Method
GET

### Endpoint
/api/v1/models/{model_id}

### Path Parameters
- **model_id** (string) - Required - The unique identifier of the model to retrieve (e.g., `sarvam-30b`).

### Response
#### Success Response (200)
Returns a single model object with all its details.
```

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

### Create a WhatsApp Bot

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/bots.md

Use this endpoint to create a new bot. Specify the bot's name, type (e.g., 'whatsapp'), and a system prompt to guide its behavior. Ensure your API key has 'bots:write' scope.

```bash
curl https://api.callmissed.com/api/v1/bots \
  -H "Authorization: Bearer cm_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Support Bot",
    "type": "whatsapp",
    "system_prompt": "You are a friendly support agent for Acme. Keep replies under 3 sentences."
  }'
```

```json
{
  "id": "b1f2c3d4-5678-90ab-cdef-1234567890ab",
  "tenant_id": "a0b1c2d3-4455-6677-8899-aabbccddeeff",
  "name": "Support Bot",
  "type": "whatsapp",
  "system_prompt": "You are a friendly support agent for Acme. Keep replies under 3 sentences.",
  "config": null,
  "is_active": true,
  "created_at": "2026-06-06T12:00:00Z",
  "updated_at": "2026-06-06T12:00:00Z",
  "conversation_count": 0
}
```

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

### Get AI Draft Reply

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/conversations.md

Generate a suggested reply from the AI without sending it to the customer. This is useful for agent co-pilots.

```http
POST /api/v1/conversations/:id/ai-draft
```

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

### Access Frontier Models

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/models-frontier.md

Use the full model ID with the /v1/chat/completions endpoint. Example shows accessing a model with a user prompt.

```python
response = client.chat.completions.create(
    model="auto",
    messages=[{"role": "user", "content": "Your prompt here"}]
)
```

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

### Usekimi-k2.5 for Fast LLM Inference

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/models.md

When the fast inference tier 'kimi-k2.5-fast' is under maintenance, fall back to the 'kimi-k2.5' model for chat completions.

```python
response = client.chat.completions.create(
    model="kimi-k2.5",
    messages=[{"role": "user", "content": "Hello"}]
)
```

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

### Get Voice Session Transcript

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-sessions-api.md

Retrieve the transcript for a voice session. Supports JSON, plain text, or SubRip subtitle formats.

```bash
curl "https://api.callmissed.com/v1/voice/sessions/{id}/transcript?format=json" \
  -H "Authorization: Bearer cm_your_api_key"
```

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

### Send Image Input to Vision-Capable Model

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/anthropic-api.md

This example demonstrates how to send an image along with text to a vision-capable model. Replace '<base64>' with your actual base64 encoded image data and 'cm_your_key' with your API key. The 'supports_vision' flag in the model listing indicates compatibility.

```bash
curl -X POST https://api.callmissed.com/v1/messages \
  -H "x-api-key: cm_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4.6",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": [{"type": "text", "text": "What is in this image?"},{"type": "image", "source": {"type": "base64", "media_type": "image/png", "data": "<base64>"}}]}]}'
```

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

### Get Specific Voice Session

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-sessions-api.md

Retrieve details for a specific voice session using its ID. Excludes sensitive creation tokens.

```bash
curl https://api.callmissed.com/v1/voice/sessions/{id} \
  -H "Authorization: Bearer cm_your_api_key"
```

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

### Get Model Access Information

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/model-access.md

Use this cURL command to fetch the model access information for all plan tiers. No authentication is required.

```bash
curl https://api.callmissed.com/api/v1/models/access
```

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

### Access Model Information with Python

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/model-access.md

This Python script demonstrates how to retrieve and parse model access data. It shows how to list free LLM models and check if a specific model is available on the free plan.

```python
import requests

resp = requests.get("https://api.callmissed.com/api/v1/models/access")
data = resp.json()

# List free LLM models
free_llms = data["plans"]["free"]["by_category"]["llm"]
print("Free LLM models:", free_llms)

# Check if a model is available on free plan
model = "openai/gpt-5.4-mini"
is_free = model in data["plans"]["free"]["models"]
print(f"{model} on free plan: {is_free}")  # False
```

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

### Execute Full Call Analytics Workflow

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/call-analytics.md

Demonstrates the complete workflow using the CallAnalytics class: analyzing an audio file, asking a question about the transcript, and retrieving a summary. Ensure the CALLMISSED_API_KEY environment variable is set.

```python
import os

analytics = CallAnalytics(api_key=os.environ["CALLMISSED_API_KEY"])

analytics.analyze("/path/to/your/audio/file.mp3", language="hi-IN")
print(analytics.answer_question("Was the customer satisfied with the resolution?"))
print(analytics.get_summary())
```

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

### Anthropic API Error Format

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/anthropic-api.md

Errors returned by the Anthropic API follow a specific JSON structure. This example shows a typical authentication error.

```json
{
  "type": "error",
  "error": {
    "type": "authentication_error",
    "message": "Invalid API key"
  }
}
```

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

### Stream Messages with JavaScript

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/anthropic-api.md

This JavaScript example demonstrates how to stream responses from the Anthropic API. It iterates over the stream events, processing text deltas as they arrive.

```javascript
import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  apiKey: "cm_your_key",
  baseURL: "https://api.callmissed.com",
});

const stream = client.messages.stream({
  model: "claude-sonnet-4.6",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Tell me a short story." }],
});

for await (const event of stream) {
  if (event.type === "content_block_delta" && event.delta.type === "text_delta") {
    process.stdout.write(event.delta.text);
  }
}
```

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

### Create and Connect to a Voice Session

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/voice-sessions-api.md

Create a voice session via the API and connect to it using the LiveKit client. Handles audio subscription and transcription events.

```javascript
import { Room, RoomEvent, Track } from "livekit-client";

const session = await fetch("https://api.callmissed.com/v1/voice/sessions", {
  method: "POST",
  headers: {
    "Authorization": "Bearer cm_your_api_key",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    system_prompt: "You are a helpful assistant.",
    voice: "shubh",
    language: "en-IN",
    llm_model: "kimi-k2.5",
  }),
}).then(r => r.json());

const room = new Room();

room.on(RoomEvent.TrackSubscribed, (track) => {
  if (track.kind === Track.Kind.Audio) {
    document.body.appendChild(track.attach());
  }
});

room.on(RoomEvent.TranscriptionReceived, (segments, participant) => {
  for (const seg of segments) {
    if (!seg.final) continue;
    const who = participant?.isLocal ? "You" : "Agent";
    console.log(who + ": " + seg.text);
  }
});

await room.connect(session.ws_url, session.token);
await room.localParticipant.setMicrophoneEnabled(true);
```

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

### TypeScript Chat Completion

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/quickstart.md

Utilize the OpenAI TypeScript SDK for chat completions. Install the SDK, set your API key and base URL, and make the request.

```typescript
import OpenAI from "openai";
const client = new OpenAI({ apiKey: "cm_your_api_key", baseURL: "https://api.callmissed.com/v1" });
const response = await client.chat.completions.create({
  model: "sarvam-30b",
  messages: [{ role: "user", content: "Hello in Hindi" }],
});
console.log(response.choices[0].message.content);
```

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

### Create Payment Order

Source: https://github.com/callmissed/callmissed-docs/blob/main/docs/payments.md

Initiates a payment order for either a credit top-up or a plan upgrade. The response contains details needed to proceed with the payment via the Cashfree SDK.

```APIDOC
## POST /api/v1/payments/create-order

### Description
Initiates a payment order for credit top-up or plan upgrade.

### Method
POST

### Endpoint
/api/v1/payments/create-order

### Parameters
#### Request Body
- **type** (string) - Required - Specifies the type of order. Can be 'credit_topup' or 'plan_upgrade'.
- **credits** (integer) - Optional - The number of credits to purchase (required if type is 'credit_topup').
- **plan** (string) - Optional - The plan to upgrade to (required if type is 'plan_upgrade', e.g., 'pro').

### Request Example
```json
{
  "type": "credit_topup",
  "credits": 5000
}
```

### Response
#### Success Response (200)
- **order_id** (string) - The internal order ID.
- **cf_order_id** (string) - The Cashfree order ID.
- **payment_session_id** (string) - The session ID for Cashfree checkout.
- **amount** (float) - The total amount for the order.
- **currency** (string) - The currency of the amount.
- **env** (string) - The environment for the Cashfree checkout (e.g., 'production').

#### Response Example
```json
{
  "order_id": "cm_9f2a1c4e6b8d0a13",
  "cf_order_id": "CF...",
  "payment_session_id": "session_...",
  "amount": 5000.0,
  "currency": "INR",
  "env": "production"
}
```
```