### Install OpenAI SDK

Source: https://docs.eachlabs.ai/sdks/javascript

Install the required dependency to interact with the OpenAI-compatible API.

```bash
npm install openai
```

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

### Install OpenAI SDK

Source: https://docs.eachlabs.ai/sense/quickstart

Install the necessary OpenAI SDK for your project using pip for Python or npm for JavaScript.

```bash
pip install openai
```

```bash
npm install openai
```

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

### Chat Completions Integration Examples

Source: https://docs.eachlabs.ai/sense/endpoints/chat-completions

Examples showing how to interact with the chat completions API using cURL, Python, and JavaScript.

```bash
curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: YOUR_API_KEY" \
  -d '{
    "messages": [{"role": "user", "content": "Generate a sunset over mountains"}],
    "stream": true,
    "session_id": "my-session",
    "mode": "max"
  }'
```

```python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)

response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{"role": "user", "content": "Generate a sunset over mountains"}],
    stream=False
)

print(response.choices[0].message.content)
```

```javascript
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://eachsense-agent.core.eachlabs.run/v1",
});

const response = await client.chat.completions.create({
  model: "eachsense/beta",
  messages: [{ role: "user", content: "Generate a sunset over mountains" }],
  stream: false,
});

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

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

### Install OpenAI Python SDK

Source: https://docs.eachlabs.ai/sdks/python

Install the necessary Python package for interacting with OpenAI-compatible APIs.

```bash
pip install openai
```

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

### 404 Not Found Error Example

Source: https://docs.eachlabs.ai/errors

Returned when the requested resource does not exist.

```json
{
  "error": "Failed to fetch model: model not found"
}
```

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

### Configure a Model Step

Source: https://docs.eachlabs.ai/workflows/concepts/workflow-structure

Example of a step configured to invoke an AI model.

```json
{
  "id": "generate_image",
  "type": "model",
  "model": "flux-1-1-pro",
  "params": {
    "prompt": "{{inputs.prompt}}",
    "aspect_ratio": "16:9"
  }
}
```

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

### Streaming SSE Events with EachLabs Extension

Source: https://docs.eachlabs.ai/llms.txt

Example of Server-Sent Events (SSE) with OpenAI chat.completion.chunk format and an EachLabs extension for generation responses. Includes example data for an image generation.

```text
data: {"id":"chatcmpl-123","object":"chat.completion.chunk","choices":[{"delta":{"content":""}}],"eachlabs":{"type":"generation_response","url":"https://cdn.example.com/image.png","generations":["..."],"model":"flux-2-max","execution_time_ms":8500}}

data: [DONE]
```

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

### GET /v1/webhooks

Source: https://docs.eachlabs.ai/api/overview

List all configured webhooks.

```APIDOC
## GET /v1/webhooks

### Description
List all webhooks associated with your account.

### Method
GET

### Endpoint
/v1/webhooks
```

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

### Fetch Workflow Details

Source: https://docs.eachlabs.ai/workflows/endpoints/get-workflow

Examples for retrieving workflow information using an API key for authentication.

```bash
curl https://workflows.eachlabs.run/api/v1/workflows/50741f40-8621-4d46-8a91-dff4d873be98 \
    -H "X-API-Key: YOUR_API_KEY"
```

```python
import requests

response = requests.get(
    "https://workflows.eachlabs.run/api/v1/workflows/text-to-image-generator",
    headers={"X-API-Key": "YOUR_API_KEY"}
)
workflow = response.json()
print(f"{workflow['name']} | v{workflow['latest_version_id']}")
```

```javascript
const response = await fetch(
  "https://workflows.eachlabs.run/api/v1/workflows/text-to-image-generator",
  { headers: { "X-API-Key": "YOUR_API_KEY" } }
);
const workflow = await response.json();
console.log(`${workflow.name} | v${workflow.latest_version_id}`);
```

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

### List Webhooks Implementation

Source: https://docs.eachlabs.ai/api/webhooks/list-webhooks

Examples for fetching a list of webhooks using cURL, Python, and JavaScript.

```bash
curl "https://api.eachlabs.ai/v1/webhooks?limit=10" \
  -H "X-API-Key: YOUR_API_KEY"
```

```python
import requests

response = requests.get(
    "https://api.eachlabs.ai/v1/webhooks",
    params={"limit": 10},
    headers={"X-API-Key": "YOUR_API_KEY"}
)

data = response.json()
for webhook in data["webhooks"]:
    print(f"{webhook['execution_id']} → {webhook['url']}")
```

```javascript
const response = await fetch(
  "https://api.eachlabs.ai/v1/webhooks?limit=10",
  { headers: { "X-API-Key": "YOUR_API_KEY" } }
);
const data = await response.json();

data.webhooks.forEach((wh) => {
  console.log(`${wh.execution_id} → ${wh.url}`);
});
```

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

### Tool chaining examples

Source: https://docs.eachlabs.ai/sense/tools

Common patterns for chaining multiple tools to achieve complex workflows.

```text
1. search_models("portrait generation")     → finds nano-banana-pro
2. get_model_details("nano-banana-pro")     → gets input schema
3. execute_model("nano-banana-pro", {...})  → generates image [TERMINAL]
```

```text
1. vision_preprocessor([image_url])         → analyzes image
2. search_models("remove background")       → finds model
3. execute_model("eachlabs-bg-remover-v1")  → removes bg [TERMINAL]
```

```text
1. search_models("portrait generation")     → finds models
2. search_models("image to video")          → finds more models
3. create_workflow({...})                   → creates workflow
4. trigger_workflow(...)                    → starts execution
5. check_execution(...)                    → monitors progress [TERMINAL]
```

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

### Streaming Response Example

Source: https://docs.eachlabs.ai/sense/endpoints/chat-completions

This example shows the Server-Sent Events (SSE) format for streaming responses, including content chunks and completion signals. Ensure your client is configured to handle SSE.

```text
data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","model":"eachsense/beta","choices":[{"index":0,"delta":{"content":"I'll generate..."},"finish_reason":null}],"eachlabs":{"type":"text_response"}}

data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","model":"eachsense/beta","choices":[{"index":0,"delta":{},"finish_reason":"stop"}],"eachlabs":{"type":"generation_response","url":"https://..."}}

data: [DONE]
```

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

### List models with filters

Source: https://docs.eachlabs.ai/api/models/list-models

Examples of how to query the models endpoint with name filtering and pagination parameters.

```bash
curl "https://api.eachlabs.ai/v1/models?name=flux&limit=10"
```

```python
import requests

response = requests.get(
    "https://api.eachlabs.ai/v1/models",
    params={"name": "flux", "limit": 10}
)

models = response.json()
for model in models:
    print(f"{model['slug']} | {model['title']}")
```

```javascript
const response = await fetch(
  "https://api.eachlabs.ai/v1/models?name=flux&limit=10"
);
const models = await response.json();

models.forEach((model) => {
  console.log(`${model.slug} | ${model.title}`);
});
```

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

### Interact with each::sense using OpenAI SDK

Source: https://docs.eachlabs.ai/llms.txt

Examples for non-streaming, streaming, eco-mode, and multi-turn interactions with the eachsense/beta model.

```python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)

# Non-streaming
response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{"role": "user", "content": "Generate a sunset landscape"}],
    stream=False
)
print(response.generations)

# Streaming
stream = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{"role": "user", "content": "Generate a sunset landscape"}],
    stream=True
)
for chunk in stream:
    eachlabs_data = chunk.model_extra.get("eachlabs")
    if eachlabs_data:
        event_type = eachlabs_data.get("type")
        if event_type == "generation_response":
            print(eachlabs_data["generations"])

# Eco mode
response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{"role": "user", "content": "Generate a quick sketch"}],
    extra_body={"mode": "eco"},
    stream=False
)

# Multi-turn with session
response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{"role": "user", "content": "Now make it wider"}],
    extra_body={"session_id": "my-session"},
    stream=False
)
```

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

### Fetch Model Information (Python)

Source: https://docs.eachlabs.ai/api/models/get-model

This Python script uses the requests library to get AI model details. Ensure you have the library installed and replace 'YOUR_API_KEY' with your actual key.

```python
import requests

response = requests.get(
    "https://api.eachlabs.ai/v1/model",
    params={"slug": "flux-1-1-pro"},
    headers={"X-API-Key": "YOUR_API_KEY"}
)

model = response.json()
print(f"Model: {model['title']}")
print(f"Input schema: {model['request_schema']}")
```

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

### Getting Your API Key

Source: https://docs.eachlabs.ai/authentication

Instructions on how to obtain an API key from the eachlabs.ai dashboard.

```APIDOC
## Getting Your API Key

1. Sign in to [eachlabs.ai](https://eachlabs.ai)
2. Navigate to **Settings** > **API Keys**
3. Click **Create API Key**
4. Copy and store your key securely

<Warning>
  Your API key grants full access to your account. Never expose it in client-side code, public repositories, or logs.
</Warning>
```

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

### each::sense (OpenAI SDK)

Source: https://docs.eachlabs.ai/llms.txt

This section demonstrates how to use the EachLabs OpenAI-compatible SDK for making requests. It includes examples for both non-streaming and streaming responses.

```APIDOC
## each::sense (OpenAI SDK)

### Description
Use the EachLabs OpenAI-compatible SDK to interact with the `eachsense/beta` model. This SDK simplifies making API calls for tasks like image generation.

### Installation
```bash
npm install openai
```

### Usage
```javascript
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://eachsense-agent.core.eachlabs.run/v1"
});

// Non-streaming example
const response = await client.chat.completions.create({
  model: "eachsense/beta",
  messages: [{ role: "user", content: "Generate a sunset landscape" }],
  stream: false
});
console.log(response.generations);

// Streaming example
const stream = await client.chat.completions.create({
  model: "eachsense/beta",
  messages: [{ role: "user", content: "Generate a sunset landscape" }],
  stream: true
});
for await (const chunk of stream) {
  const eachlabs = chunk.eachlabs;
  if (eachlabs?.type === "generation_response") {
    console.log(eachlabs.generations);
  }
}
```

### Parameters
- **apiKey** (string): Your unique EachLabs API key.
- **baseURL** (string): The base URL for the EachLabs API endpoint.
- **model** (string): The model to use for generation (e.g., `eachsense/beta`).
- **messages** (array): An array of message objects, typically with `role` and `content`.
- **stream** (boolean): Whether to stream the response. Defaults to `false`.
```

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

### List Models via Python

Source: https://docs.eachlabs.ai/sense/endpoints/models

Use the OpenAI Python client to list models. Ensure you have the `openai` library installed and replace YOUR_API_KEY and the base URL if necessary.

```python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)

models = client.models.list()
for model in models.data:
    print(f"{model.id} | owned by {model.owned_by}")
```

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

### Fetch Categories Implementation

Source: https://docs.eachlabs.ai/workflows/endpoints/categories

Examples for authenticating and retrieving category data using cURL, Python, and JavaScript.

```bash
curl https://workflows.eachlabs.run/api/v1/categories \
  -H "X-API-Key: YOUR_API_KEY"
```

```python
import requests

response = requests.get(
    "https://workflows.eachlabs.run/api/v1/categories",
    headers={"X-API-Key": "YOUR_API_KEY"}
)

for cat in response.json()["categories"]:
    print(f"{cat['slug']}: {cat['label']}")
```

```javascript
const response = await fetch(
  "https://workflows.eachlabs.run/api/v1/categories",
  { headers: { "X-API-Key": "YOUR_API_KEY" } }
);
const { categories } = await response.json();
categories.forEach((c) => console.log(`${c.slug}: ${c.label}`));
```

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

### GET /v1/models

Source: https://docs.eachlabs.ai/api/overview

Retrieve a list of all available AI models.

```APIDOC
## GET /v1/models

### Description
List all available AI models accessible through the platform.

### Method
GET

### Endpoint
/v1/models
```

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

### Choice Step with Simple Condition

Source: https://docs.eachlabs.ai/llms.txt

Route execution based on a condition. This example uses a simple string match operator to determine which branch to execute.

```json
{
  "id": "step2",
  "type": "choice",
  "condition": {
    "expression": "$.step1.primary",
    "operator": "string_matches",
    "value": ".mp4"
  },
  "condition_met_branch": {
    "name": "condition_met",
    "step_id": "step2_condition_met",
    "steps": [
      {"id": "step2_branch_0", "type": "model", "model": "topaz-upscale-video", "params": {"video": "{{step1.primary}}", "scale": 2}}
    ]
  },
  "default_branch": {
    "name": "default",
    "step_id": "step2_default",
    "steps": [
      {"id": "step2_default", "type": "model", "model": "topaz-upscale-image", "params": {"image_url": "{{step1.primary}}", "scale": 2}}
    ]
  }
}
```

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

### Fetch Model Information (JavaScript)

Source: https://docs.eachlabs.ai/api/models/get-model

This JavaScript example demonstrates fetching AI model data using the fetch API. It requires an API key for authentication.

```javascript
const response = await fetch(
    "https://api.eachlabs.ai/v1/model?slug=flux-1-1-pro",
    {
      headers: { "X-API-Key": "YOUR_API_KEY" },
    }
);
const model = await response.json();
console.log(`Model: ${model.title}`);
```

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

### HTTP Step Configuration

Source: https://docs.eachlabs.ai/llms.txt

Example configuration for an HTTP step within a workflow, including URL, method, headers, body, and authentication.

```APIDOC
### HTTP Steps

```json
{
  "id": "call_api",
  "type": "http",
  "url": "https://api.example.com/process",
  "method": "POST",
  "headers": {"Content-Type": "application/json"},
  "body": {"data": "{{step1.primary}}"},
  "auth": {"type": "bearer", "token": "{{inputs.api_token}}"},
  "timeout": 30000
}
```

**Auth types**: `basic` (username/password), `bearer` (token), `api_key`.
```

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

### Generate Image with each::sense (Python OpenAI SDK)

Source: https://docs.eachlabs.ai/llms.txt

This Python snippet demonstrates how to generate media using the each::sense AI agent via the OpenAI SDK. Ensure you have the 'openai' library installed and configure the API key and base URL.

```python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)

response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{"role": "user", "content": "Generate a cinematic sunset landscape in 16:9"}],
    stream=False
)

print(response.generations)  # List of media URLs
```

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

### List Models via JavaScript

Source: https://docs.eachlabs.ai/sense/endpoints/models

Use the OpenAI JavaScript client to list models. Ensure you have the `openai` library installed and replace YOUR_API_KEY and the base URL if necessary.

```javascript
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://eachsense-agent.core.eachlabs.run/v1",
});

const models = await client.models.list();
models.data.forEach((m) => console.log(`${m.id} | ${m.owned_by}`));
```

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

### Perform Chat Completions

Source: https://docs.eachlabs.ai/sense/overview

Examples for generating content using the OpenAI SDK or direct cURL requests.

```python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_EACHLABS_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)

response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{"role": "user", "content": "Generate a portrait photo"}],
    stream=True
)

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

```bash
curl -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: YOUR_API_KEY" \
  -d '{
    "messages": [{"role": "user", "content": "Generate a portrait photo"}],
    "stream": false
  }'
```

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

### Create Workflow via Natural Language with Python

Source: https://docs.eachlabs.ai/sense/examples

Create a workflow by describing it in natural language. This example demonstrates creating a workflow for image generation and animation.

```python
import requests
import json

URL = "https://eachsense-agent.core.eachlabs.run/v1/chat/completions"
HEADERS = {"Content-Type": "application/json", "X-API-Key": "YOUR_API_KEY"}

# Create workflow
response = requests.post(URL, headers=HEADERS, json={
    "messages": [{
        "role": "user",
        "content": "Create a workflow: 1) Generate portrait from description, 2) Animate into 5s video"
    }],
    "stream": True
}, stream=True)

for line in response.iter_lines():
    if not line:
        continue
    line = line.decode("utf-8")
    if line.startswith("data: ") and line[6:] != "[DONE]":
        event = json.loads(line[6:])
        if event["type"] == "workflow_created":
            print(f"Workflow created: {event['workflow_id']}")
        elif event["type"] == "text_response":
            print(event["content"])
```

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

### Python Step Configuration

Source: https://docs.eachlabs.ai/llms.txt

Example configuration for a Python step within a workflow, allowing custom code execution.

```APIDOC
### Python Steps

```json
{
  "id": "transform",
  "type": "python",
  "code": "result = inputs['text'].upper()",
  "inputs": {"text": "{{step1.primary}}"}
}
```
```

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

### Workflow Creation Events

Source: https://docs.eachlabs.ai/sense/workflows

These are example events emitted by the agent during workflow creation, indicating status updates and the final creation confirmation.

```text
data: {"type":"status","message":"Creating workflow..."}
data: {"type":"workflow_created","workflow_id":"wf_abc123","version_id":"v1","steps_count":3}
data: {"type":"text_response","content":"I've created your workflow!"}
data: {"type":"complete","status":"ok"}
data: [DONE]
```

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

### Python SSE Chat Stream with OpenAI SDK

Source: https://docs.eachlabs.ai/sense/streaming/implementation

Utilize the OpenAI SDK in Python to stream chat completions. This example shows how to configure the client with a custom base URL and API key, and how to access both standard content and custom 'eachlabs' extensions from the stream chunks.

```python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)

stream = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{"role": "user", "content": "Generate a portrait"}],
    stream=True
)

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

    # Access each::labs extensions
    if hasattr(chunk, "model_extra") and chunk.model_extra:
        eachlabs = chunk.model_extra.get("eachlabs", {})
        if eachlabs.get("type") == "generation_response":
            print(f"\nGenerated: {eachlabs.get('url')}")
```

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

### Get Session Memory Response

Source: https://docs.eachlabs.ai/sense/endpoints/sessions-memory

Example JSON response containing conversation history and total exchange count.

```json
{
  "session_id": "my-session",
  "conversation_history": [
    {
      "timestamp": "2024-01-15T10:30:00Z",
      "user_prompt": "Generate a portrait",
      "chatbot_response": "Here's your portrait!",
      "generated_media_urls": ["https://storage.eachlabs.ai/xxx.png"]
    }
  ],
  "total_exchanges": 1,
  "generated_media_urls": ["https://storage.eachlabs.ai/xxx.png"]
}
```

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

### Initialize OpenAI Client

Source: https://docs.eachlabs.ai/sdks/javascript

Configure the OpenAI client to point to the each::sense base URL.

```javascript
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://eachsense-agent.core.eachlabs.run/v1",
});
```

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

### Initialize OpenAI-compatible client

Source: https://docs.eachlabs.ai/llm-router/quickstart

Configure the client to point to the each::labs base URL using your API key.

```python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_EACHLABS_API_KEY",
    base_url="https://api.eachlabs.ai/v1"
)
```

```typescript
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_EACHLABS_API_KEY",
  baseURL: "https://api.eachlabs.ai/v1",
});
```

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

### Fetch Prediction Status

Source: https://docs.eachlabs.ai/api/predictions/get-prediction

Examples for retrieving prediction status using cURL, Python, and JavaScript. The Python and JavaScript examples implement a polling loop to wait for completion.

```bash
curl https://api.eachlabs.ai/v1/prediction/abc123-def456-ghi789 \
  -H "X-API-Key: YOUR_API_KEY"
```

```python
import requests
import time

prediction_id = "abc123-def456-ghi789"

while True:
    response = requests.get(
        f"https://api.eachlabs.ai/v1/prediction/{prediction_id}",
        headers={"X-API-Key": "YOUR_API_KEY"}
    )
    data = response.json()

    if data["status"] in ("success", "failed", "cancelled"):
        break

    print(f"Status: {data['status']}...")
    time.sleep(2)

if data["status"] == "success":
    print(f"Output: {data['output']}")
else:
    print(f"Failed: {data.get('logs')}")
```

```javascript
async function waitForPrediction(predictionId) {
  while (true) {
    const response = await fetch(
      `https://api.eachlabs.ai/v1/prediction/${predictionId}`,
      { headers: { "X-API-Key": "YOUR_API_KEY" } }
    );
    const data = await response.json();

    if (["success", "failed", "cancelled"].includes(data.status)) {
      return data;
    }

    console.log(`Status: ${data.status}...`);
    await new Promise((r) => setTimeout(r, 2000));
  }
}

const result = await waitForPrediction("abc123-def456-ghi789");
console.log(`Output: ${result.output}`);
```

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

### Get Model Details with cURL

Source: https://docs.eachlabs.ai/api/overview

Use this command to retrieve details for a specific AI model, such as its input parameters and capabilities. Replace YOUR_API_KEY with your actual API key.

```bash
curl https://api.eachlabs.ai/v1/model?slug=flux-1-1-pro \
  -H "X-API-Key: YOUR_API_KEY"
```

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

### Trigger Workflow Execution (JavaScript)

Source: https://docs.eachlabs.ai/workflows/endpoints/trigger-workflow

This JavaScript example uses the fetch API to trigger a workflow. It configures the POST request with appropriate headers and a JSON body containing inputs and a webhook URL, then logs the execution ID.

```javascript
const response = await fetch(
    "https://workflows.eachlabs.run/api/v1/WF_ID/trigger",
    {
      method: "POST",
      headers: {
        "Content-Type": "application/json",
        "X-API-Key": "YOUR_API_KEY",
      },
      body: JSON.stringify({
        inputs: { prompt: "A majestic mountain landscape" },
        webhook_url: "https://your-app.com/webhooks/workflow",
      }),
    }
  );
  const { execution_id } = await response.json();
  console.log(`Execution ID: ${execution_id}`);
```

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

### Error Response Examples

Source: https://docs.eachlabs.ai/sense/endpoints/chat-completions

These examples illustrate common error response bodies for different HTTP status codes. Use these to identify and handle issues like missing API keys, invalid credentials, or rate limits.

```json
{"detail": "API key is required."}
```

```json
{"detail": "Invalid API key."}
```

```json
{"detail": [{"loc": ["body","messages"], "msg": "field required"}]}
```

```json
{"detail": "Rate limit exceeded."}
```

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

### GET /categories

Source: https://docs.eachlabs.ai/llms.txt

Retrieve a list of available workflow categories.

```APIDOC
## GET /categories

### Description
Retrieve a list of available workflow categories.

### Method
GET

### Endpoint
/categories
```

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

### Create a New Workflow

Source: https://docs.eachlabs.ai/sense/endpoints/workflow-builder

Demonstrates how to initialize a new AI workflow using cURL or Python. Requires an API key in the request header.

```bash
curl -X POST https://eachsense-agent.core.eachlabs.run/workflow \
  -H "Content-Type: application/json" \
  -H "X-API-Key: YOUR_API_KEY" \
  -d '{
    "message": "Create a workflow that generates a portrait image then animates it into a 5-second video",
    "stream": true
  }'
```

```python
import requests

response = requests.post(
    "https://eachsense-agent.core.eachlabs.run/workflow",
    headers={
        "Content-Type": "application/json",
        "X-API-Key": "YOUR_API_KEY"
    },
    json={
        "message": "Create a workflow that generates a portrait then animates it",
        "stream": False
    }
)
print(response.json())
```

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

### GET /v1/webhooks

Source: https://docs.eachlabs.ai/api/webhooks/list-webhooks

Retrieve a paginated list of webhooks for your organization.

```APIDOC
## GET https://api.eachlabs.ai/v1/webhooks

### Description
Retrieve a paginated list of webhooks for your organization.

### Method
GET

### Endpoint
https://api.eachlabs.ai/v1/webhooks

### Parameters
#### Query Parameters
- **limit** (integer) - Optional - Max webhooks to return (1–100), default 50
- **offset** (integer) - Optional - Number of webhooks to skip, default 0

### Request Example
curl "https://api.eachlabs.ai/v1/webhooks?limit=10" -H "X-API-Key: YOUR_API_KEY"

### Response
#### Success Response (200)
- **webhooks** (array) - List of webhook objects
- **webhooks[].execution_id** (string | null) - Associated execution ID
- **webhooks[].url** (string) - Target URL
- **webhooks[].request** (string) - Original request payload
- **webhooks[].headers** (object) - Headers included in the webhook
- **webhooks[].source** (string) - Service that triggered the webhook
- **webhooks[].created_at** (string) - ISO 8601 timestamp
- **limit** (integer) - Applied limit
- **offset** (integer) - Applied offset

#### Response Example
{
  "webhooks": [
    {
      "execution_id": "abc123-def456-ghi789",
      "url": "https://api.example.com/webhook",
      "request": "{\"event\":\"prediction.completed\"}",
      "headers": {
        "Content-Type": "application/json"
      },
      "source": "api-gateway",
      "created_at": "2025-12-14T10:30:00Z"
    }
  ],
  "limit": 50,
  "offset": 0
}
```

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

### GET /workflows/{id}

Source: https://docs.eachlabs.ai/llms.txt

Retrieve details for a specific workflow by its ID.

```APIDOC
## GET /workflows/{id}

### Description
Retrieve details for a specific workflow by its ID.

### Method
GET

### Endpoint
/workflows/{id}

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

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

### Providing Initial Request Details

Source: https://docs.eachlabs.ai/sense/clarifications

Include all necessary parameters in the initial request to prevent the agent from needing to ask for clarification.

```json
{
  "messages": [{
    "role": "user",
    "content": "Generate an anime-style portrait, 1:1 aspect ratio, cyberpunk background, neon lighting"
  }]
}
```

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

### Get Public Workflow

Source: https://docs.eachlabs.ai/workflows/overview

Retrieves details of a publicly shared workflow.

```APIDOC
## GET /public/@{nickname}/workflows/{slug}/versions/{versionID}

### Description
Get a public workflow.

### Method
GET

### Endpoint
/public/@{nickname}/workflows/{slug}/versions/{versionID}

### Parameters
#### Path Parameters
- **nickname** (string) - Required - The nickname of the workflow owner.
- **slug** (string) - Required - The slug of the workflow.
- **versionID** (string) - Required - The identifier for the workflow version.
```

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

### Parameter References (Template Syntax)

Source: https://docs.eachlabs.ai/llms.txt

Explains how to reference inputs and step outputs using double brace `{{ }}` template syntax and condition expressions using `$` syntax.

```APIDOC
## Parameter References (Template Syntax)

Use `{{double braces}}` to reference inputs and step outputs:

| Reference | Description | Example |
|-----------|-------------|---------|
| `{{inputs.field}}` | Workflow input | `{{inputs.prompt}}` |
| `{{step_id.output}}` | Full output of a step | `{{generate.output}}` |
| `{{step_id.primary}}` | Primary/first result | `{{generate.primary}}` |
| `{{branch_step_id.primary}}` | Output from a parallel branch step | `{{step1_branch_0.primary}}` |

Templates can be embedded in strings: `"A portrait of {{inputs.name}} in {{inputs.style}} style"`

Condition expressions use `$` syntax: `$.step_id.primary`, `$.step_id.output`, `$.step_id.output.field`, `$.inputs.field`
```

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

### Get Execution Details

Source: https://docs.eachlabs.ai/workflows/overview

Retrieves the details of a specific workflow execution.

```APIDOC
## GET /executions/{executionID}

### Description
Get execution details.

### Method
GET

### Endpoint
/executions/{executionID}

### Parameters
#### Path Parameters
- **executionID** (string) - Required - The unique identifier of the execution.
```

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

### Get Workflow Details

Source: https://docs.eachlabs.ai/workflows/overview

Retrieves the details of a specific workflow by its ID.

```APIDOC
## GET /workflows/{id}

### Description
Get workflow details.

### Method
GET

### Endpoint
/workflows/{id}

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

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

### GET /v1/webhooks/{execution_id}

Source: https://docs.eachlabs.ai/api/overview

Retrieve details for a specific webhook execution.

```APIDOC
## GET /v1/webhooks/{execution_id}

### Description
Get detailed information regarding a specific webhook execution.

### Method
GET

### Endpoint
/v1/webhooks/{execution_id}

### Parameters
#### Path Parameters
- **execution_id** (string) - Required - The unique identifier for the webhook execution.
```

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

### GET /v1/model

Source: https://docs.eachlabs.ai/api/overview

Retrieve detailed information for a specific AI model.

```APIDOC
## GET /v1/model

### Description
Get specific details for a model, including input requirements and pricing.

### Method
GET

### Endpoint
/v1/model

### Parameters
#### Query Parameters
- **slug** (string) - Required - The unique identifier for the model.
```

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

### Update Events

Source: https://docs.eachlabs.ai/sense/endpoints/workflow-builder

Example of streaming events returned during a workflow update process.

```text
data: {"type":"workflow_fetched","workflow_name":"portrait-to-video","existing_steps":3}

data: {"type":"workflow_built","steps_count":4,"definition":{...}}

data: {"type":"workflow_updated","success":true,"workflow_id":"wf_abc123","version_id":"v2"}

data: {"type":"complete","status":"ok"}

data: [DONE]
```

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

### 500 Internal Server Error Example

Source: https://docs.eachlabs.ai/errors

Returned when an unexpected server-side error occurs.

```json
{
  "error": "Failed to fetch models: internal error"
}
```

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

### 403 Forbidden Error Example

Source: https://docs.eachlabs.ai/errors

Returned when an operation is not allowed, such as modifying a locked resource.

```json
{
  "error": "Workflow is locked and cannot be modified"
}
```

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

### Generate media using the each::sense AI agent

Source: https://docs.eachlabs.ai/

Utilize the OpenAI-compatible client to interact with the each::sense agent for automated model selection and media generation.

```python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)

response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{"role": "user", "content": "Generate a cinematic landscape at golden hour"}],
)
```

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

### 400 Bad Request Error Example

Source: https://docs.eachlabs.ai/errors

Returned when request parameters or the body are invalid.

```json
{
  "error": "slug parameter is required"
}
```

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

### Get Session Memory

Source: https://docs.eachlabs.ai/sense/endpoints/sessions-memory

Retrieve the conversation history and memory for a specific session.

```APIDOC
## GET /memory

### Description
Retrieve the conversation history and memory for a specific session.

### Method
GET

### Endpoint
https://eachsense-agent.core.eachlabs.run/memory

### Query Parameters
- **session_id** (string) - Required - Session ID to retrieve

### Response
#### Success Response (200)
- **session_id** (string) - The ID of the session.
- **conversation_history** (array) - An array of conversation objects, each containing timestamp, user prompt, chatbot response, and generated media URLs.
- **total_exchanges** (integer) - The total number of exchanges in the conversation.
- **generated_media_urls** (array) - URLs of media generated during the session.

### Request Example
```json
{
  "session_id": "my-session"
}
```

### Response Example
```json
{
  "session_id": "my-session",
  "conversation_history": [
    {
      "timestamp": "2024-01-15T10:30:00Z",
      "user_prompt": "Generate a portrait",
      "chatbot_response": "Here's your portrait!",
      "generated_media_urls": ["https://storage.eachlabs.ai/xxx.png"]
    }
  ],
  "total_exchanges": 1,
  "generated_media_urls": ["https://storage.eachlabs.ai/xxx.png"]
}
```
```

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

### POST /workflows

Source: https://docs.eachlabs.ai/llms.txt

Create a new workflow instance.

```APIDOC
## POST /workflows

### Description
Create a new workflow instance.

### Method
POST

### Endpoint
/workflows
```

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

### Get Prediction Endpoint

Source: https://docs.eachlabs.ai/api/predictions/get-prediction

The base URL for retrieving prediction status and results.

```http
GET https://api.eachlabs.ai/v1/prediction/{id}
```

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

### Execution Events

Source: https://docs.eachlabs.ai/sense/streaming/event-types

Events related to the start, progress, and completion of workflow executions.

```APIDOC
## Execution Events

### `execution_started`

### Request Example
```json
{ "type": "execution_started", "execution_id": "exec_xyz", "workflow_id": "wf_abc123" }
```

### `execution_progress`

### Request Example
```json
{
  "type": "execution_progress",
  "step_id": "step2",
  "step_status": "completed",
  "model": "kling-2-1-image-to-video",
  "output": "https://storage.eachlabs.ai/video.mp4",
  "completed_steps": 2,
  "total_steps": 4,
  "progress_percent": 50
}
```

### `execution_completed`

### Request Example
```json
{
  "type": "execution_completed",
  "execution_id": "exec_xyz",
  "status": "completed",
  "output": "https://storage.eachlabs.ai/final.mp4",
  "all_outputs": { "step1": "...", "step2": "...", "step3": "..." },
  "total_time_ms": 45000
}
```
```

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

### List Models via cURL

Source: https://docs.eachlabs.ai/sense/endpoints/models

Use this cURL command to list available models. Replace YOUR_API_KEY with your actual API key.

```bash
curl https://eachsense-agent.core.eachlabs.run/v1/models \
  -H "X-API-Key: YOUR_API_KEY"
```

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

### Execution Events

Source: https://docs.eachlabs.ai/sense/streaming/event-types

Events tracking the start, progress, and completion of workflow executions.

```json
{ "type": "execution_started", "execution_id": "exec_xyz", "workflow_id": "wf_abc123" }
```

```json
{
  "type": "execution_progress",
  "step_id": "step2",
  "step_status": "completed",
  "model": "kling-2-1-image-to-video",
  "output": "https://storage.eachlabs.ai/video.mp4",
  "completed_steps": 2,
  "total_steps": 4,
  "progress_percent": 50
}
```

```json
{
  "type": "execution_completed",
  "execution_id": "exec_xyz",
  "status": "completed",
  "output": "https://storage.eachlabs.ai/final.mp4",
  "all_outputs": { "step1": "...", "step2": "...", "step3": "..." },
  "total_time_ms": 45000
}
```

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

### Enable SSE Streaming with cURL

Source: https://docs.eachlabs.ai/sense/streaming/overview

Use this cURL command to send a POST request to the completions endpoint and enable streaming. Set `stream: true` in the request body. Ensure your API key is included in the headers.

```bash
curl -N -X POST https://eachsense-agent.core.eachlabs.run/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key: YOUR_API_KEY" \
  -d '{
    "messages": [{"role": "user", "content": "Generate a portrait"}],
    "stream": true
  }'
```

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

### Streaming Response Events

Source: https://docs.eachlabs.ai/sense/endpoints/workflow-builder

Example of Server-Sent Events (SSE) returned when streaming is enabled.

```text
data: {"type":"status","message":"Creating workflow..."}

data: {"type":"workflow_created","workflow_id":"wf_abc123","version_id":"v1","input_schema":{...},"steps_count":3}

data: {"type":"text_response","content":"I've created your workflow!"}

data: {"type":"complete","status":"ok","workflow_id":"wf_abc123"}

data: [DONE]
```

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

### 429 Rate Limited Error Example

Source: https://docs.eachlabs.ai/errors

Returned when the request frequency exceeds allowed limits.

```json
{
  "error": "Rate limit exceeded. Please retry after 60 seconds."
}
```

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

### Configure AI agent client for each::sense

Source: https://docs.eachlabs.ai/

Point any OpenAI-compatible framework to the each::sense base URL to enable agent-based workflows.

```python
# Works with LangChain, CrewAI, AutoGen, or any OpenAI-compatible client
client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)
```

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

### each::workflows - Get execution details

Source: https://docs.eachlabs.ai/llms.txt

Retrieves detailed information about a specific workflow execution.

```APIDOC
## GET /executions/{executionID}

### Description
Get execution details.

### Method
GET

### Endpoint
`https://workflows.eachlabs.run/api/v1/executions/{executionID}`

### Parameters
#### Path Parameters
- **executionID** (string) - Required - The unique identifier of the execution.
```

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

### Interact with each::sense using OpenAI SDK

Source: https://docs.eachlabs.ai/llms.txt

Configure the OpenAI client with the EachLabs base URL and perform streaming or non-streaming chat completions.

```javascript
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://eachsense-agent.core.eachlabs.run/v1"
});

// Non-streaming
const response = await client.chat.completions.create({
  model: "eachsense/beta",
  messages: [{ role: "user", content: "Generate a sunset landscape" }],
  stream: false
});
console.log(response.generations);

// Streaming
const stream = await client.chat.completions.create({
  model: "eachsense/beta",
  messages: [{ role: "user", content: "Generate a sunset landscape" }],
  stream: true
});
for await (const chunk of stream) {
  const eachlabs = chunk.eachlabs;
  if (eachlabs?.type === "generation_response") {
    console.log(eachlabs.generations);
  }
}
```

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

### Initiate Chat Completion with Python

Source: https://docs.eachlabs.ai/llm-router/models

Use the client.chat.completions.create method to send a request to a specific model. Ensure the model parameter follows the provider/model-name format.

```python
response = client.chat.completions.create(
    model="openai/gpt-4o",  # any model ID from the tables below
    messages=[{"role": "user", "content": "Hello!"}]
)
```

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

### each::workflows - Get workflow

Source: https://docs.eachlabs.ai/llms.txt

Retrieves details of a specific workflow using its ID or slug.

```APIDOC
## GET /workflows/{id}

### Description
Get workflow details.

### Method
GET

### Endpoint
`https://workflows.eachlabs.run/api/v1/workflows/{id}`

### Parameters
#### Path Parameters
- **id** (string) - Required - The unique identifier (UUID or slug) of the workflow.
```

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

### GET /v1/prediction/{id}

Source: https://docs.eachlabs.ai/api/predictions/get-prediction

Retrieve the status and results of a model prediction by providing its ID.

```APIDOC
## GET /v1/prediction/{id}

### Description
Retrieve the status and results of a model prediction.

### Method
GET

### Endpoint
https://api.eachlabs.ai/v1/prediction/{id}

### Parameters
#### Path Parameters
- **id** (string) - Required - Prediction ID returned from [Create Prediction](/api/predictions/create-prediction)

### Request Example
```bash
curl https://api.eachlabs.ai/v1/prediction/abc123-def456-ghi789 \
  -H "X-API-Key: YOUR_API_KEY"
```

### Response
#### Success Response (200)
- **id** (string) - Prediction ID
- **input** (object) - Input parameters used
- **status** (string) - `starting`, `processing`, `success`, `failed`, or `cancelled`
- **output** (string | array | object) - Prediction output (type depends on model)
- **logs** (string | null) - Execution logs
- **metrics.predict_time** (number) - Processing time in seconds
- **metrics.cost** (number) - Cost in USD
- **urls.cancel** (string) - URL to cancel the prediction
- **urls.get** (string) - URL to re-fetch this prediction

#### Response Example
```json
{
  "id": "abc123-def456-ghi789",
  "input": {
    "prompt": "A beautiful sunset over the ocean with vibrant colors",
    "aspect_ratio": "16:9"
  },
  "status": "success",
  "output": "https://storage.example.com/predictions/abc123/image.jpg",
  "logs": null,
  "metrics": {
    "predict_time": 12.5,
    "cost": 0.05
  },
  "urls": {
    "cancel": "https://api.eachlabs.ai/v1/prediction/abc123-def456-ghi789/cancel",
    "get": "https://api.eachlabs.ai/v1/prediction/abc123-def456-ghi789"
  }
}
```

### Error Responses
| Status | Body                                 | Description           |
| ------ | ------------------------------------ | --------------------- |
| `404`  | `{"error": "Prediction not found"}`  | Invalid prediction ID |
| `500`  | `{"error": "Internal server error"}` | Server error          |
```