curl --request POST \
  --url https://asteragents.com/api/kb/manage \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "name": "Product Documentation",
  "embeddingModel": "openai:text-embedding-3-small"
}
'

{
  "success": true,
  "knowledgeBase": {
    "id": 1,
    "name": "Product Documentation",
    "description": null,
    "embeddingModel": "openai:text-embedding-3-small",
    "embeddingDimensions": 1536,
    "extractionModel": null,
    "extractionSchema": null,
    "source": null,
    "trigger": null,
    "createdAt": "2024-01-15T10:30:00.000Z",
    "updatedAt": "2024-01-15T10:30:00.000Z",
    "userId": "user_2ABC123DEF",
    "fileCount": 0,
    "status": "empty"
  }
}

Knowledge Bases API

Create a knowledge base

Create a new knowledge base in the authenticated user’s organization.

Required fields:

name — must be unique within the organization
embeddingModel — the model used to generate embeddings (e.g., openai:text-embedding-3-small)

Optional features:

source — marks the KB as integration-managed (disables manual file upload in UI)
trigger — automatically runs an agent when new files are added
extractionModel / extractionSchema — structured data extraction from uploaded files

POST

/

kb

/

manage

curl --request POST \
  --url https://asteragents.com/api/kb/manage \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "name": "Product Documentation",
  "embeddingModel": "openai:text-embedding-3-small"
}
'

{
  "success": true,
  "knowledgeBase": {
    "id": 1,
    "name": "Product Documentation",
    "description": null,
    "embeddingModel": "openai:text-embedding-3-small",
    "embeddingDimensions": 1536,
    "extractionModel": null,
    "extractionSchema": null,
    "source": null,
    "trigger": null,
    "createdAt": "2024-01-15T10:30:00.000Z",
    "updatedAt": "2024-01-15T10:30:00.000Z",
    "userId": "user_2ABC123DEF",
    "fileCount": 0,
    "status": "empty"
  }
}

Authorizations

Authorization

string

header

required

JWT token from Clerk authentication.

Must be from a user with org:admin role.

Body

application/json

name

string

required

Name of the knowledge base (must be unique within the org)

Example:

"Product Documentation"

embeddingModel

string

required

Embedding model identifier (e.g., openai:text-embedding-3-small)

Example:

"openai:text-embedding-3-small"

description

string | null

Optional description

Example:

"All product docs and user guides"

extractionModel

string | null

Optional model for structured data extraction from files

extractionSchema

object | null

JSON schema defining what structured data to extract from files

source

string | null

Integration source identifier. When set, the KB becomes integration-managed — manual file upload/deletion is disabled in the UI, but API access remains unrestricted.

Example:

"salesforce"

trigger

object

Trigger configuration that automatically runs an agent when new files are added. Set enabled: true with an agentId and prompt to activate.

Show child attributes

Response

Knowledge base created successfully

success

boolean

required

knowledgeBase

object

required

Show child attributes