List attention items
GET /api/attention-items
Returns all attention items for the tenant. Attention items are created automatically when the AI needs human intervention — for example, when reply validation fails, when a customer requests a human agent, or when the AI detects a sensitive situation it cannot handle. Use this endpoint to build a queue of items requiring human review.
Authorizations
Section titled “Authorizations ”Responses
Section titled “ Responses ”List of attention items, including both open and resolved items. Filter client-side by status if needed. An API key restricted to a brand or workspace receives the items of the workspaces its restriction covers.
List of attention items requiring human review.
An attention item flags a conversation that needs human review. These are created automatically when the AI is unable to handle a situation — for example, when reply validation fails after 3 attempts, when a customer explicitly requests a human agent, or when the AI detects a sensitive situation it cannot handle.
object
Unique identifier for the attention item.
Example
ai1a2b3c-5678-9abc-def0-1234567890abThe tenant this attention item belongs to.
Example
a0b1c2d3-4567-89ab-cdef-0123456789abThe conversation that triggered this attention item.
Example
c1a2b3c4-5678-9abc-def0-1234567890abThe category of attention required. Common values: ‘validation_failed’ (AI reply failed fact-checking), ‘human_requested’ (customer asked for a human), ‘escalation’ (AI detected a sensitive issue).
Example
validation_failedDetailed explanation of why the attention item was created.
Example
Reply validation failed after 3 attempts. The AI was unable to verify pricing information against the knowledge base.Current status of the attention item. ‘open’ means unresolved, ‘resolved’ means a human has reviewed and addressed it.
Example
openTimestamp when the attention item was created.
Example
2026-03-25T11:42:00.000ZTimestamp when the attention item was resolved. Null while still open.
A conversationId filter that is not an id, or, from an API key not restricted to a brand or workspace, a workspaceId filter that is not an id (Invalid filter).
Standard error response returned by all endpoints on failure.
object
A human-readable error message describing what went wrong.
Example
Conversation not foundA machine-readable reason on some refusals, for example ‘key_lock’ when an API key’s brand or workspace restriction does not cover the action.
Example
key_lockOn a 403 for an API key without a scope the endpoint requires: the scopes the key lacks.
Example
[ "mcp:write"]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.
Example
11223344-5566-4788-99aa-bbccddeeff00On 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.
On the same 422: the tool names the connection does not declare.
Example
[ "get_orders"]On the same 422: the tools the connection declares.
Example
[ "get_order", "search_orders"]