Create a KB entry

POST

/api/workspaces/{workspaceId}/kb-entries

Production

Creates a new knowledge base entry in the specified workspace. Manually created entries are immediately set to ‘approved’ status and become available to the AI. Provide a clear question-answer pair and relevant topic tags to help the AI match this entry to customer inquiries.

Authorizations

bearerAuth

Parameters

Path Parameters

workspaceId

required

string format: uuid

The workspace to create the entry in.

Request Body

object

question

required

The question or topic this entry addresses.

string

>= 1 characters

Example

What are your delivery times to Germany?

answer

required

The answer to the question.

string

>= 1 characters

Example

Standard delivery to Germany takes 3-5 business days. Express delivery (next business day) is available for EUR 12.50.

sourceType

required

How this entry was created. Use ‘manual’ for human-authored entries, ‘import’ for API-imported entries.

string

>= 1 characters

Example

manual

topics

Tags that categorize this entry. Helps the AI match relevant entries to conversations.

Array<string>

Example

[
  "shipping",
  "delivery",
  "germany"
]

sourceConversationId

Optional reference to the conversation that inspired this entry.

string

Example

c1a2b3c4-5678-9abc-def0-1234567890ab

confidence

Confidence score between 0 and 1. Typically set by the learning loop, not manually.

number

<= 1

Example

Responses

201

The entry was created.

A knowledge base (KB) entry is a question-answer pair that the AI uses to respond to customer inquiries. Entries can be created manually, imported via the API, or proposed automatically by the learning loop after resolved conversations. Proposed entries require human approval before the AI can use them.

object

required

Unique identifier for the KB entry.

string format: uuid

Example

kb1a2b3c-5678-9abc-def0-1234567890ab

tenantId

required

The tenant this entry belongs to.

string format: uuid

Example

a0b1c2d3-4567-89ab-cdef-0123456789ab

workspaceId

required

The workspace this entry is scoped to. The AI only uses entries from the workspace that matches the conversation.

string format: uuid

Example

w1a2b3c4-5678-9abc-def0-1234567890ab

question

required

The question or topic this entry addresses. This is what the AI matches against when a customer asks something.

string

Example

What are your delivery times to Germany?

answer

required

The answer the AI should use when this entry is matched. Can include specific details, links, and instructions.

string

Example

Standard delivery to Germany takes 3-5 business days. Express delivery (next business day) is available for an additional fee of EUR 12.50.

topics

required

Tags that categorize this entry. Used for filtering and to help the AI determine relevance.

Array<string>

Example

[
  "shipping",
  "delivery",
  "germany"
]

sourceType

required

How the entry came to be. manual is an entry written by a person, approved when it is created. extracted is a proposal the learning loop drafted from a resolved conversation. The API stores the value sent, and an entry created with a value other than manual starts as pending_review.

string

Example

manual

confidence

Confidence score between 0 and 1, set by the learning loop when proposing entries. Higher values indicate the AI is more confident the entry is accurate. Null for manually created entries.

number | null

Example

0.92

status

required

Where the entry stands. approved: in use, the assistant draws on it. pending_review: created for review, through the API with a sourceType other than manual; it is not used until it is approved. proposed: drafted by oHallo, for example by the learning loop from a resolved conversation; it is not used until it is approved, and a proposal that changes an existing entry is applied to that entry when it is approved. rejected: declined in review. archived: no longer in use, kept for the record.

string

Allowed values: pending_review approved rejected archived proposed

Example

approved

usageCount

required

Number of times this entry has been referenced by the AI when responding to conversations.

integer

Example

lastUsedAt

Timestamp of the last time the AI referenced this entry. Null if never used.

string | null format: date-time

Example

2026-03-25T14:20:00.000Z

createdAt

required

Timestamp when the entry was created.

string format: date-time

Example

2026-02-01T12:00:00.000Z

updatedAt

required

Timestamp when the entry was last updated.

string format: date-time

Example

2026-03-15T10:30:00.000Z

409

The workspace already has an entry with this question. Each question is held by one entry of the workspace, compared without case or surrounding spaces; a rejected entry holds none. code is duplicate_question and entryId names the entry that holds it.

Standard error response returned by all endpoints on failure.

object

error

required

A human-readable error message describing what went wrong.

string

Example

Conversation not found

code

A machine-readable reason on some refusals, for example ‘key_lock’ when an API key’s brand or workspace restriction does not cover the action.

string

Example

key_lock

missing_scopes

On a 403 for an API key without a scope the endpoint requires: the scopes the key lacks.

Array<string>

Example

[
  "mcp:write"
]

entryId

On a 409 with code duplicate_question or duplicate_rule_text: the knowledge base entry or policy that already holds the question or rule text.

string format: uuid

Example

11223344-5566-4788-99aa-bbccddeeff00

connectionId

On a 422 about one MCP connection a request names, from an assignment or an agent create or update (a connection that does not exist, one listed twice, or one that does not declare a tool named): the connection, as the request named it.

string

undeclaredTools

On the same 422: the tool names the connection does not declare.

Array<string>

Example

[
  "get_orders"
]

declaredTools

On the same 422: the tools the connection declares.

Array<string>

Example

[
  "get_order",
  "search_orders"
]

422

Validation error. One or more required fields are missing or invalid.

Standard error response returned by all endpoints on failure.

object

error

required

A human-readable error message describing what went wrong.

string

Example

Conversation not found

code

A machine-readable reason on some refusals, for example ‘key_lock’ when an API key’s brand or workspace restriction does not cover the action.

string

Example

key_lock

missing_scopes

On a 403 for an API key without a scope the endpoint requires: the scopes the key lacks.

Array<string>

Example

[
  "mcp:write"
]

entryId

On a 409 with code duplicate_question or duplicate_rule_text: the knowledge base entry or policy that already holds the question or rule text.

string format: uuid

Example

11223344-5566-4788-99aa-bbccddeeff00

connectionId

string

undeclaredTools

On the same 422: the tool names the connection does not declare.

Array<string>

Example

[
  "get_orders"
]

declaredTools

On the same 422: the tools the connection declares.

Array<string>

Example

[
  "get_order",
  "search_orders"
]