Searching and Filtering
The IntelCenter API allows you to filter and refine results using query parameters across all datasets.
Two types of search are available:
- Keyword Search — precise, filter-driven queries
- Natural Language Search — conceptual, AI-assisted retrieval
Keyword Search Example
Code
Natural Language Search Example
Code
Supported Query Parameters (Keyword Search)
| Parameter | Description |
|---|---|
| search | Full text search across indexed fields |
| entity | Filter by associated entity |
| country | Filter by primary country |
| date_from | Filter results on or after a date (YYYY-MM-DD) |
| date_to | Filter results on or before a date (YYYY-MM-DD) |
| page | Page number (default: 1) |
| per_page | Results per page (capped by your API key’s limit) |
| sort | Optional sorting (advanced usage) |
Supported Query Parameters (Natural Language Search)
| Parameter | Description |
|---|---|
| search | Natural language query |
| entity | Optional entity filter |
| country | Optional country filter |
| source_component | Filter by content type |
| top_k | Controls retrieval depth |
| page | Page number |
| per_page | Results per page |
Pagination
Search results are paginated using page and per_page.
Example:
Code
Response fields include:
curPage— current pagenextPage/prevPage— navigationperPage— results per pageitemsTotal— total matching recordspageTotal— total number of pages
If no pagination parameters are provided, default values will be used.
Note: The per_page value may be limited based on your API key tier.
Combining Filters
Multiple filters can be combined in a single request:
Code
You can also combine full text search with filters:
Code
Natural Language Search also supports filters:
Code
Dataset Compatibility
Keyword search is supported across all datasets:
- /v1/icd/videos
- /v1/icdb/videos
- /v1/icd/images
- /v1/icdb/images
Natural Language Search endpoints:
- /v1/icd/nl-search
- /v1/icdb/nl-search
Choosing the Right Search Method
Use Keyword Search when:
- You know exact terms (entity names, keywords)
- You need precise filtering
- You want predictable, structured results
Use Natural Language Search when:
- You are exploring a topic
- You need conceptual matching
- You want results across transcripts, analysis, and metadata
Performance Notes
- Indexed fields such as
search,entity, andcountryare optimized for fast querying - Pagination improves performance and response times
- Combining filters reduces result size and improves efficiency
- Natural Language Search may be slightly slower due to deeper retrieval
Default Behavior
- All query parameters are optional
- Results are returned in descending chronological order
- Parameter names use snake_case
Usage Limits
Search requests are subject to rate limits and quotas based on your API key.
- Requests per minute may be limited
- Daily and monthly quotas may apply
Natural Language Search may have different rate limits depending on your API key tier.
If limits are exceeded, the API will return a 429 Too Many Requests response.
MCP Usage
When using MCP endpoints, these same parameters are available but passed as structured inputs instead of query strings.
Tips
- Use keyword search for precision
- Use natural language search for discovery
- Combine filters to narrow results
- Use pagination to control response size
- Adjust
top_kin NL search to balance depth vs performance
