Re-embed All Knowledges

Used to re-process and re-embed all knowledge base entries. This is useful when updating embedding models, refreshing the knowledge base index, or applying new configurations to existing content.

API Endpoint

Property	Value
Request Method	POST
Request URL	https://api.seliseblocks.com/kb/reembed

Request

Request Example

curl -X POST 'https://api.seliseblocks.com/kb/reembed' \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
    "project_key": "YOUR_PROJECT_KEY",
    "agent_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "kbs": [
      {
        "kb_id": "kb_001",
        "type": "file",
        "source": "document.pdf"
      },
      {
        "kb_id": "kb_002",
        "type": "text",
        "source": "raw_text_content"
      }
    ]
  }'

Request Headers

Field	Type	Required	Description
accept	string	Yes	Accepted response format. Use application/json
Content-Type	application/json	Yes	Data type, must be application/json.

Request Body

Request Body Schema

{
  "project_key": "string",
  "agent_id": "string",
  "kbs": [
    {
      "additionalProp1": {}
    }
  ]
}

Request Body Parameters

Field	Type	Required	Description
project_key	string	Yes	The project key for your project.
agent_id	string	Yes	UUID of the AI agent whose knowledge base will be re-embedded.
kbs	array	Yes	Array of knowledge base entry objects to re-embed.

Knowledge Base Entry Object

Each entry in the kbs array is an object containing information about a specific knowledge base item. The structure may vary depending on the knowledge type:

Field	Type	Description
kb_id	string	Unique identifier of the knowledge base entry.
type	string	Type of knowledge (e.g., file, text, qa, link).
source	string	Source identifier or content reference.

note

When to Re-embed

• After changing the embedding model in RAG configuration
• When updating to a newer version of an embedding model
• To refresh the vector index with improved embeddings
• After making significant changes to chunking strategies

warning

Re-embedding can be a resource-intensive operation, especially for large knowledge bases. The process may take considerable time depending on:

• The number of knowledge base entries
• The size of the content
• The complexity of the embedding model
• Current system load

Response

Success Response (200 OK)

Returns an object containing the re-embedding status.

{
  "is_success": true,
  "detail": "Knowledge base entries re-embedded successfully. Processed 150 entries.",
  "error": {}
}

Response Fields

Field	Type	Description
is_success	boolean	Indicates whether the re-embedding was successful.
detail	string	Success or failure message with additional context.
error	object	Error details if the operation failed (empty if successful).

Error Response (422 Unprocessable Entity)

Returns validation error details when the request body is invalid.

{
  "detail": [
    {
      "loc": [
        "body",
        "kbs"
      ],
      "msg": "field required",
      "type": "value_error.missing"
    }
  ]
}

Error Response Fields

Field	Type	Description
detail	array	Array of validation error objects.
loc	array	Location of the error in the request (e.g., body field).
msg	string	Human-readable error message.
type	string	Error type identifier.

Error Codes

Status Code	Description	Response Type
200	Successful Response	Success
400	Bad Request - Invalid knowledge base data	Bad Request
422	Validation Error - Invalid request parameters	Unprocessable Entity
500	Internal Server Error - Re-embedding failed	Server Error