> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.twelvelabs.io/v1.3/api-reference/knowledge-stores/create/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.twelvelabs.io/_mcp/server.

# Create a knowledge store

POST https://api.twelvelabs.io/v1.3/knowledge-stores
Content-Type: application/json

This method creates a knowledge store.

Provide a name. Optionally include a description, a metadata map, and an `ingestion_config` object that controls how content added to the store is processed. The `ingestion_config` object is immutable after creation.


Reference: https://docs.twelvelabs.io/api-reference/knowledge-stores/create

## Authentication

- `x-api-key` header (required) — Your API key. You can find your API key on the API Keys page.

## Request

### Body (application/json)

This endpoint expects an object.

- `name` (string, required) — The name of the knowledge store.
- `ingestion_config` (IngestionConfig, optional) — Configuration that controls how content added to the knowledge store is processed.
- `description` (string, optional) — An optional description of the knowledge store.
- `metadata` (map from string to KnowledgeStoreMetadataValue, optional) — Custom metadata for the knowledge store, as user-defined key-value pairs. Up to 10 pairs, keys up to 128 characters, string values up to 2048 characters. Keys are strings; values can be a string, a number, a boolean, or an array of strings. A nested object, an array containing anything other than strings, and a null value are rejected. An integer must fit in 53 bits (-9007199254740991 to 9007199254740991). Send a wider integer, or an identifier that must be preserved verbatim, as a string.

## Response

### 201

The knowledge store has been successfully created.

- `_id` (string, optional) — The unique identifier of the knowledge store.
- `name` (string, optional) — The name of the knowledge store.
- `description` (string, optional) — An optional description of the knowledge store.
- `ingestion_config` (IngestionConfig, optional) — The ingestion configuration set when this knowledge store was created. Immutable.
- `item_count` (integer, optional) — The number of items in the knowledge store.
- `created_at` (string, optional) — The date and time when the knowledge store was created, in the RFC 3339 format.
- `updated_at` (string, optional) — The date and time when the knowledge store was last updated, in the RFC 3339 format.
- `metadata` (map from string to KnowledgeStoreMetadataValue, optional) — Custom metadata for the knowledge store. Keys are strings; each value is a string, a number, a boolean, or an array of strings.

## Errors

### 400 Bad Request Error

The request has failed.

- `code` (string, optional) — A string representing the code associated with the error. See the [Error codes](/v1.3/api-reference/error-codes) page for details.
- `message` (string, optional) — A human-readable string describing the error, intended to be suitable for display in a user interface.

## Types

### IngestionConfig

Configuration that controls how content added to the knowledge store is processed.

- `enrichment_config` (EnrichmentConfig, required) — Metadata enrichment configuration. The `type` field selects one variant: - `type: json_schema` - Structured extraction conforming to a provided JSON Schema. - `type: description` - Natural-language instructions that the platform converts into a schema internally.

### KnowledgeStoreMetadataValue

A single custom-metadata value: a string, a number, a boolean, or an array of strings. The value keeps the JSON type sent; a nested object, an array containing anything other than strings, and a null value are rejected. An integer must fit in 53 bits (-9007199254740991 to 9007199254740991). A wider integer, or an identifier that must be preserved verbatim, must be sent as a string.

### EnrichmentConfig

Metadata enrichment configuration. The `type` field selects one variant: - `type: json_schema` - Structured extraction conforming to a provided JSON Schema. - `type: description` - Natural-language instructions that the platform converts into a schema internally.

- `type`: `json_schema` (json_schema)
  - `json_schema` (EnrichmentConfigJsonSchemaJsonSchema, required) — A JSON Schema (draft 2020-12) that defines the structure of metadata to extract from each video shot. The platform accepts only the JSON Schema keywords listed below; unknown keywords return a `422` error. Keep schemas focused — deeply nested or overly complex schemas may degrade extraction quality. **Supported keywords** | Category | Keywords | |---|---| | Core | `type`, `title`, `description`, `enum` | | Object | `properties`, `required`, `additionalProperties` | | Array | `items`, `prefixItems`, `minItems`, `maxItems` | | Number | `minimum`, `maximum` | | String | `format` | Notes: - The root `type` keyword must be `"object"`. - Every entry under the `properties` keyword (including nested ones) must include a `description` field. This text guides extraction quality. - The `required` keyword behaves as in standard JSON Schema. - The `additionalProperties` keyword accepts a boolean value (`true` or `false`). Use `false` to enforce strict shapes. The platform does not support schema-valued forms. - **Unknown keywords return `422` — they are not silently ignored.** If you generate schemas programmatically (for example, using `pydantic.model_json_schema()`), strip annotative keywords such as `default`, `examples`, and `readOnly` before submitting. - **The platform does not support nullable fields.** This includes both `nullable: true` and `type: ["string", "null"]`. To express an optional field, omit it from the `required` array — the platform omits the field from the result when no value is present. **Not supported (returns `422`)** - Schema composition: `anyOf`, `allOf`, `oneOf`, `not` - Conditional schemas: `if` / `then` / `else` - Property dependencies: `dependentSchemas`, `dependentRequired` - Object size constraints: `minProperties`, `maxProperties` - String constraints: `pattern`, `minLength`, `maxLength` - Number constraints: `exclusiveMinimum`, `exclusiveMaximum`, `multipleOf` - Value constraints: `const` - Null handling: `nullable`, `type` arrays (e.g., `["string", "null"]`) - Annotative keywords: `default`, `examples`, `readOnly`, `writeOnly` - References and reuse: `$ref`, `$defs`, `definitions` - Any other keyword not listed under "Supported keywords" above
- `type`: `description` (description)
  - `description` (string, required) — A natural-language description of what knowledge should be extracted from the videos. The platform converts this into a JSON Schema internally.

### EnrichmentConfigJsonSchemaJsonSchema

A JSON Schema (draft 2020-12) that defines the structure of metadata to extract from each video shot. The platform accepts only the JSON Schema keywords listed below; unknown keywords return a `422` error. Keep schemas focused — deeply nested or overly complex schemas may degrade extraction quality. **Supported keywords** | Category | Keywords | |---|---| | Core | `type`, `title`, `description`, `enum` | | Object | `properties`, `required`, `additionalProperties` | | Array | `items`, `prefixItems`, `minItems`, `maxItems` | | Number | `minimum`, `maximum` | | String | `format` | Notes: - The root `type` keyword must be `"object"`. - Every entry under the `properties` keyword (including nested ones) must include a `description` field. This text guides extraction quality. - The `required` keyword behaves as in standard JSON Schema. - The `additionalProperties` keyword accepts a boolean value (`true` or `false`). Use `false` to enforce strict shapes. The platform does not support schema-valued forms. - **Unknown keywords return `422` — they are not silently ignored.** If you generate schemas programmatically (for example, using `pydantic.model_json_schema()`), strip annotative keywords such as `default`, `examples`, and `readOnly` before submitting. - **The platform does not support nullable fields.** This includes both `nullable: true` and `type: ["string", "null"]`. To express an optional field, omit it from the `required` array — the platform omits the field from the result when no value is present. **Not supported (returns `422`)** - Schema composition: `anyOf`, `allOf`, `oneOf`, `not` - Conditional schemas: `if` / `then` / `else` - Property dependencies: `dependentSchemas`, `dependentRequired` - Object size constraints: `minProperties`, `maxProperties` - String constraints: `pattern`, `minLength`, `maxLength` - Number constraints: `exclusiveMinimum`, `exclusiveMaximum`, `multipleOf` - Value constraints: `const` - Null handling: `nullable`, `type` arrays (e.g., `["string", "null"]`) - Annotative keywords: `default`, `examples`, `readOnly`, `writeOnly` - References and reuse: `$ref`, `$defs`, `definitions` - Any other keyword not listed under "Supported keywords" above

- `type` (enum, required) — The top-level type of the schema. Must be `"object"`. Primitive and array top-level types are not supported. The platform uses this object structure to map extracted metadata to named fields.
  - Allowed values: `object`
- `properties` (map from string to any, required) — A map of property names to their JSON Schema definitions. Each property must include a field named `description`. The platform uses this field to guide extraction quality. Omitting it returns a `422` error.

## Examples

**Request**

```json
{
  "name": "Product Demo Analysis"
}
```

**Response**

```json
{
  "_id": "ks_069e9869-1ea3-7481-8000-dae72bf6be6e",
  "name": "Product Demo Analysis",
  "description": "Knowledge store for product demo videos",
  "ingestion_config": {
    "enrichment_config": {
      "type": "json_schema",
      "json_schema": {
        "type": "object",
        "properties": {
          "topic": {
            "type": "string",
            "description": "The main topic discussed in this shot"
          },
          "sentiment": {
            "type": "string",
            "enum": [
              "positive",
              "negative",
              "neutral"
            ],
            "description": "The overall sentiment of the shot"
          }
        },
        "required": [
          "topic"
        ]
      }
    }
  },
  "item_count": 5,
  "created_at": "2024-01-15T09:30:00Z",
  "updated_at": "2024-01-15T09:30:00Z",
  "metadata": {}
}
```

**SDK Code**

```python
import requests

url = "https://api.twelvelabs.io/v1.3/knowledge-stores"

payload = { "name": "Product Demo Analysis" }
headers = {
    "x-api-key": "<apiKey>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.json())
```

```javascript
const url = 'https://api.twelvelabs.io/v1.3/knowledge-stores';
const options = {
  method: 'POST',
  headers: {'x-api-key': '<apiKey>', 'Content-Type': 'application/json'},
  body: '{"name":"Product Demo Analysis"}'
};

try {
  const response = await fetch(url, options);
  const data = await response.json();
  console.log(data);
} catch (error) {
  console.error(error);
}
```

```go
package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.twelvelabs.io/v1.3/knowledge-stores"

	payload := strings.NewReader("{\n  \"name\": \"Product Demo Analysis\"\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("x-api-key", "<apiKey>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(res)
	fmt.Println(string(body))

}
```

```ruby
require 'uri'
require 'net/http'

url = URI("https://api.twelvelabs.io/v1.3/knowledge-stores")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["x-api-key"] = '<apiKey>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"name\": \"Product Demo Analysis\"\n}"

response = http.request(request)
puts response.read_body
```

```java
import com.mashape.unirest.http.HttpResponse;
import com.mashape.unirest.http.Unirest;

HttpResponse<String> response = Unirest.post("https://api.twelvelabs.io/v1.3/knowledge-stores")
  .header("x-api-key", "<apiKey>")
  .header("Content-Type", "application/json")
  .body("{\n  \"name\": \"Product Demo Analysis\"\n}")
  .asString();
```

```php
<?php
require_once('vendor/autoload.php');

$client = new \GuzzleHttp\Client();

$response = $client->request('POST', 'https://api.twelvelabs.io/v1.3/knowledge-stores', [
  'body' => '{
  "name": "Product Demo Analysis"
}',
  'headers' => [
    'Content-Type' => 'application/json',
    'x-api-key' => '<apiKey>',
  ],
]);

echo $response->getBody();
```

```csharp
using RestSharp;

var client = new RestClient("https://api.twelvelabs.io/v1.3/knowledge-stores");
var request = new RestRequest(Method.POST);
request.AddHeader("x-api-key", "<apiKey>");
request.AddHeader("Content-Type", "application/json");
request.AddParameter("application/json", "{\n  \"name\": \"Product Demo Analysis\"\n}", ParameterType.RequestBody);
IRestResponse response = client.Execute(request);
```

```swift
import Foundation

let headers = [
  "x-api-key": "<apiKey>",
  "Content-Type": "application/json"
]
let parameters = ["name": "Product Demo Analysis"] as [String : Any]

let postData = JSONSerialization.data(withJSONObject: parameters, options: [])

let request = NSMutableURLRequest(url: NSURL(string: "https://api.twelvelabs.io/v1.3/knowledge-stores")! as URL,
                                        cachePolicy: .useProtocolCachePolicy,
                                    timeoutInterval: 10.0)
request.httpMethod = "POST"
request.allHTTPHeaderFields = headers
request.httpBody = postData as Data

let session = URLSession.shared
let dataTask = session.dataTask(with: request as URLRequest, completionHandler: { (data, response, error) -> Void in
  if (error != nil) {
    print(error as Any)
  } else {
    let httpResponse = response as? HTTPURLResponse
    print(httpResponse)
  }
})

dataTask.resume()
```