Search Collection

Search within collections using fulltext, semantic, or hybrid search capabilities

POST /api/v1/collections/search

cURL

curl -X POST "https://api.mielto.com/api/v1/collections/search" \
  -H "X-API-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "How to configure API authentication",
    "collection_id": "col_123456789",
    "search_type": "hybrid",
    "max_results": 10
  }'

Response

{
  "results": [
    {
      "content": "API authentication can be configured using either API keys or Bearer tokens. To use API keys, include the X-API-Key header in your requests...",
      "score": 0.95,
      "metadata": {
        "document_title": "Authentication Guide",
        "section": "API Keys",
        "page": 1
      },
      "content_id": "content_123456",
      "source": "authentication-guide.pdf"
    },
    {
      "content": "When configuring authentication, ensure you store your API keys securely and never expose them in client-side code...",
      "score": 0.87,
      "metadata": {
        "document_title": "Security Best Practices",
        "section": "API Security",
        "page": 3
      },
      "content_id": "content_789012",
      "source": "security-guide.pdf"
    }
  ],
  "total_results": 2,
  "query": "How to configure API authentication",
  "search_type": "hybrid",
  "collection_id": "col_123456789",
  "execution_time_ms": 45.2
}

Search Collection

Search within a collection using knowledge search with support for different search types including fulltext, semantic, and hybrid search.

Headers

Authorization · string

Bearer token for authentication

X-API-Key · string

API key for authentication (alternative to Authorization header)

X-Workspace-Id · string

Optional. Derived from API Key, but can be specified if you have access to multiple workspaces

Request Body

query · string · required

Search query string

collection_id · string · required

ID of the collection to search in

search_type · string · default: "hybrid"

Type of search to perform (vector, keyword, hybrid)

max_results · integer · default: "10"

Maximum number of results to return (1-100)

filters · object

Additional filters to apply to the search

Response

results · array

Array of search results

Show search result

content · string

The content text that matched

score · number

Relevance score (0-1)

metadata · object

Additional metadata for the result

content_id · string

ID of the source content

source · string

Source of the content

total_results · integer

Total number of results found

query · string

Original search query

search_type · string

Type of search performed

collection_id · string

Collection that was searched

execution_time_ms · number

Search execution time in milliseconds

cURL

curl -X POST "https://api.mielto.com/api/v1/collections/search" \
  -H "X-API-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "How to configure API authentication",
    "collection_id": "col_123456789",
    "search_type": "hybrid",
    "max_results": 10
  }'

Response

{
  "results": [
    {
      "content": "API authentication can be configured using either API keys or Bearer tokens. To use API keys, include the X-API-Key header in your requests...",
      "score": 0.95,
      "metadata": {
        "document_title": "Authentication Guide",
        "section": "API Keys",
        "page": 1
      },
      "content_id": "content_123456",
      "source": "authentication-guide.pdf"
    },
    {
      "content": "When configuring authentication, ensure you store your API keys securely and never expose them in client-side code...",
      "score": 0.87,
      "metadata": {
        "document_title": "Security Best Practices",
        "section": "API Security",
        "page": 3
      },
      "content_id": "content_789012",
      "source": "security-guide.pdf"
    }
  ],
  "total_results": 2,
  "query": "How to configure API authentication",
  "search_type": "hybrid",
  "collection_id": "col_123456789",
  "execution_time_ms": 45.2
}

Search types:

• vector: Semantic similarity search using embeddings
• keyword: Traditional keyword-based search
• hybrid: Combination of vector and keyword search (recommended)