client.kb
Knowledge-base documents assistants answer from. These methods are on client.kb, where client is a MoreVoice client (see the Python SDK). On AsyncMoreVoice the same methods are awaited. Each one returns the response object and raises an exception when the API answers with an error.
documents_search()
Section titled “documents_search()”Search the knowledge base. The passages that best answer query, ranked like assistants and the copilot rank them (hybrid: words and meaning). Test keys search by words only, without an embedding call.
def documents_search(self, body: _m.KbSearchParamsInput | Mapping[str, Any], *, more_voice_version: str | Unset = UNSET, idempotency_key: str | Unset = UNSET) -> _m.KbSearchResultclient.kb.documents_search() · await async_client.kb.documents_search() · POST /kb/search · API reference
Request body
Section titled “Request body”| Field | Type | Required | Description |
|---|---|---|---|
query |
string |
yes | What to look for, in any language the documents are in. |
copilot_profile_id |
string |
no | Search as this copilot profile does: its documents only. |
document_ids |
string[] |
no | Search only these documents. |
limit |
integer |
no | Passages to return (1–20, default 5). |
Returns
Section titled “Returns”KbSearchResult:
| Field | Type | Description |
|---|---|---|
object |
"kb_search_result" |
Always kb_search_result. |
livemode |
boolean |
true in live mode, false in test mode. |
query |
string |
— |
data |
object[] |
— |
semantic |
boolean |
Meaning-based retrieval took part. Test keys search by words only (no embedding call). |
Example
Section titled “Example”from morevoice import MoreVoice
client = MoreVoice() # MOREVOICE_API_KEY from the environment
kb_search_result = client.kb.documents_search({ "limit": 3, "query": "When are you open on Friday?",})print(kb_search_result)