Hybrid search across the Knowledge Base, combining semantic similarity with keyword matching. Returns top-K matched chunks with citations.

POSThttps://app.softsolz.uk/api/v1/services/smart-chat/kb/search

Recent Requests

Kept only in this browser

TimeStatusTook

Send a request with Try it to see it here.

Headers

Authorizationstringrequired

Bearer token: `Bearer sk_live_...` (or `sk_test_...` for sandbox). A login token will not work here.

Idempotency-Keystring

Optional idempotency key for safe retries.

Body Params

querystringrequired

The search query.

top_kinteger

Max chunks to return (default 6).

document_idsarray of strings

Restrict to specific document ids.

tagsarray of strings

Restrict to documents with these tags.

foldersarray of strings

Restrict to these folders.

Response

Response Body

hitsarray of objectsMatched chunks, most relevant first. `similarity` is the cosine similarity to the query (0-1) and is the field to rank or threshold on; it is null when a chunk was matched by keywords alone. `matched_keywords` is true when the chunk matched the keyword leg. `score` is an internal rank-fusion value whose magnitude carries no meaning outside a single response.

modelstring | nullEmbedding model used.

usage_tokensinteger | nullEmbedding tokens consumed by the query.

Request

curl --request POST \
  --url https://app.softsolz.uk/api/v1/services/smart-chat/kb/search \
  --header 'Authorization: Bearer sk_live_your_key' \
  --header 'Content-Type: application/json' \
  --data '{
  "query": "refund policy for late payments",
  "top_k": 6,
  "tags": [
    "policy"
  ],
  "folders": [
    "legal"
  ]
}'

Credentials

Authorization

Sent as a Bearer token and used in the samples above and in Try it. Kept only in this browser tab until you close it. Use an sk_test_ key to stay in your sandbox.

Response · 200

{
  "data": {
    "hits": [
      {
        "document_id": "d0000000-0000-0000-0000-000000000001",
        "file_name": "refund-policy.pdf",
        "page_number": 3,
        "similarity": 0.62,
        "matched_keywords": true,
        "score": 0.0328,
        "snippet": "Refunds for late payments..."
      }
    ],
    "model": "text-embedding-3-small",
    "usage_tokens": 12
  }
}