Semantic Search Cases
semantic_search_casesSearch Indian court cases using natural language questions about legal concepts, fact patterns, or doctrines, even without exact keyword matches.
Instructions
AI vector search over the roughly 2M case subset that has embeddings, out of the full 310M plus corpus. CONSUMES AI CREDITS. Prefer this over search_indian_court_cases for natural language questions about legal concepts, fact patterns or doctrines, where exact keywords will not match. Slower, a single call can take a minute or more. Real behaviour note: this endpoint's validation layer also accepts top level court, year, caseType, caseNumber, judgeName, judges, judge, fromDate and toDate fields, but the handler silently ignores all of them, only query, page, limit and filters are actually used, so this tool only exposes those. Put any filtering inside the filters object instead. Another quirk: if the cleaned query text ends up shorter than 3 characters after internal processing, the service falls back to a plain keyword search and marks the response with meta.fallbackMode = "opensearch".
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| page | No | Page number, default 1. | |
| limit | No | Results per page, 1 to 100, default 20, clamped to 100 server side. | |
| query | Yes | Natural language question or description, minimum 3 characters. | |
| filters | No | Free form filter object passed straight through to the vector store. Keys the handler actually understands downstream: court, caseType, caseYear, judgeName, caseNumber, and decisionDate as an object with $gte and or $lte sub keys for range filtering. Filters supplied here override whatever filters the service would otherwise auto extract from the query text. |