> ## Documentation Index > Fetch the complete documentation index at: https://docs.memic.ai/llms.txt > Use this file to discover all available pages before exploring further. # Unified search > Runs unified search across both unstructured documents and structured connector-backed data in your environment. Memic classifies the incoming query and intelligently routes it — natural-language questions are answered by semantic search over indexed documents, while structured queries (filters, aggregations, lookups) hit connector-backed sources. Returns ranked results with source citations plus routing metadata so you can see which path was taken. ## OpenAPI ````yaml /api-reference/openapi.json post /api/v1/search openapi: 3.1.0 info: title: Memic API description: >- Memic is a memory layer for AI applications. The Memic API lets you ingest documents, run semantic search and chat, and manage prompts — all scoped to an environment via a single API key. version: 1.0.0 servers: - url: https://api.memic.ai description: Production security: - MemicApiKey: [] tags: - name: Endpoints description: All 10 Memic API endpoints. paths: /api/v1/search: post: tags: - Endpoints summary: Unified search description: >- Runs unified search across both unstructured documents and structured connector-backed data in your environment. Memic classifies the incoming query and intelligently routes it — natural-language questions are answered by semantic search over indexed documents, while structured queries (filters, aggregations, lookups) hit connector-backed sources. Returns ranked results with source citations plus routing metadata so you can see which path was taken. operationId: search requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/SearchRequest' responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/SearchResponse' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - MemicApiKey: [] components: schemas: SearchRequest: properties: query: type: string maxLength: 1000 minLength: 1 title: Query description: The search query text examples: - What are the key findings in the cybersecurity report? project_id: anyOf: - type: string format: uuid - type: 'null' title: Project Id description: Optional project ID to limit search scope file_ids: anyOf: - items: type: string format: uuid type: array - type: 'null' title: File Ids description: Optional list of file IDs to limit search scope environment_slug: anyOf: - type: string - type: 'null' title: Environment Slug description: >- Optional environment slug to filter search ('staging' or 'production') top_k: type: integer maximum: 50 minimum: 1 title: Top K description: Number of results to return (max 50) default: 10 min_score: type: number maximum: 1 minimum: 0 title: Min Score description: Minimum similarity score threshold (0.0 to 1.0) default: 0.7 include_metadata: type: boolean title: Include Metadata description: Whether to include file and chunk metadata in results default: true connector_id: anyOf: - type: string format: uuid - type: 'null' title: Connector Id description: 'Optional: Force query to use a specific connector' force_route: anyOf: - $ref: '#/components/schemas/QueryRouteEnum' - type: 'null' description: 'Optional: Force query routing (structured, semantic, or hybrid)' metadata_filters: anyOf: - $ref: '#/components/schemas/MetadataFilters' - type: 'null' description: >- Metadata filters for vector search (reference_id, page_number, etc.). Used in lookup-based search where SQL returns filter values. type: object required: - query title: SearchRequest description: Request model for semantic search. SearchResponse: properties: query: type: string title: Query description: Original search query results: $ref: '#/components/schemas/SearchResults' description: All search results by type total_results: type: integer title: Total Results description: Total number of results returned routing: anyOf: - $ref: '#/components/schemas/RoutingInfo' - type: 'null' description: Query routing information search_time_ms: type: number title: Search Time Ms description: Search execution time in milliseconds namespace: type: string title: Namespace description: Organization namespace used for search filters_applied: additionalProperties: true type: object title: Filters Applied description: Filters applied to the search type: object required: - query - total_results - search_time_ms - namespace - filters_applied title: SearchResponse description: Response model for semantic search. example: filters_applied: project_id: d1710969-5973-4cf1-8c4e-3264158e765f namespace: 8e9fd1e0-e355-4394-9878-b59ec0fffcfd query: quarterly revenue results: semantic: - category: Security chunk_id: 123e4567-e89b-12d3-a456-426614174000 chunk_index: 5 content: The cybersecurity landscape shows increasing threats... file_id: 123e4567-e89b-12d3-a456-426614174001 file_name: Cyber_Report_2024.pdf mime_type: application/pdf page_number: 3 score: 0.92 tags: - cybersecurity - report token_count: 256 structured: columns: - description: Customer name name: customer type: varchar - name: revenue type: integer rows: - customer: Acme revenue: 50000 - customer: Beta revenue: 75000 routing: reasoning: Query combines document search with structured data route: hybrid search_time_ms: 125.5 total_results: 3 HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError QueryRouteEnum: type: string enum: - structured - semantic - hybrid title: QueryRouteEnum description: Query routing options. MetadataFilters: properties: reference_id: anyOf: - type: string - type: 'null' title: Reference Id description: Client-provided file reference ID (matches file.reference_id) reference_ids: anyOf: - items: type: string type: array - type: 'null' title: Reference Ids description: Multiple reference IDs to match (OR logic) page_number: anyOf: - type: integer minimum: 1 - type: 'null' title: Page Number description: Exact page number match page_numbers: anyOf: - items: type: integer type: array - type: 'null' title: Page Numbers description: Match any of these page numbers (OR logic) page_range: anyOf: - $ref: '#/components/schemas/PageRangeFilter' - type: 'null' description: Page number range filter (inclusive) category: anyOf: - type: string - type: 'null' title: Category description: Exact category match document_type: anyOf: - type: string - type: 'null' title: Document Type description: Exact document type match type: object title: MetadataFilters description: >- Type-safe metadata filters for vector search. Used for lookup-based search where SQL query returns filter values (e.g., reference_id, page_number) that are then applied to vector search. Example usage: # Filter by reference_id (from SQL: file_ref = 'TG_G1_Math') {"reference_id": "TG_G1_Math"} # Filter by reference_id + exact page {"reference_id": "TG_G1_Math", "page_number": 40} # Filter by reference_id + multiple pages {"reference_id": "TG_G1_Math", "page_numbers": [40, 41, 42]} # Filter by reference_id + page range {"reference_id": "TG_G1_Math", "page_range": {"gte": 40, "lte": 50}} SearchResults: properties: semantic: items: $ref: '#/components/schemas/SearchResultChunk' type: array title: Semantic description: Semantic search results from vector store default: [] structured: anyOf: - $ref: '#/components/schemas/StructuredResult' - type: 'null' description: Structured query results with schema type: object title: SearchResults description: Container for all result types. RoutingInfo: properties: route: type: string title: Route description: 'Route used: structured, semantic, or hybrid' reasoning: anyOf: - type: string - type: 'null' title: Reasoning description: Why this route was chosen connector_id: anyOf: - type: string format: uuid - type: 'null' title: Connector Id description: Connector used for structured query connector_name: anyOf: - type: string - type: 'null' title: Connector Name description: Name of connector used sql_generated: anyOf: - type: string - type: 'null' title: Sql Generated description: SQL query generated (if structured) sql_explanation: anyOf: - type: string - type: 'null' title: Sql Explanation description: Explanation of SQL query type: object required: - route title: RoutingInfo description: Information about how the query was routed. ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type type: object required: - loc - msg - type title: ValidationError PageRangeFilter: properties: gte: anyOf: - type: integer minimum: 1 - type: 'null' title: Gte description: Page number greater than or equal to lte: anyOf: - type: integer minimum: 1 - type: 'null' title: Lte description: Page number less than or equal to type: object title: PageRangeFilter description: Filter pages by range (inclusive). SearchResultChunk: properties: chunk_id: type: string format: uuid title: Chunk Id description: Database chunk ID file_id: type: string format: uuid title: File Id description: File ID the chunk belongs to file_name: type: string title: File Name description: Original filename project_id: anyOf: - type: string format: uuid - type: 'null' title: Project Id description: Project ID content: type: string title: Content description: Chunk text content score: type: number title: Score description: Similarity score (0.0 to 1.0) chunk_index: type: integer title: Chunk Index description: Index of chunk within file page_number: anyOf: - type: integer - type: 'null' title: Page Number description: Page number if available start_page: anyOf: - type: integer - type: 'null' title: Start Page description: Starting page number end_page: anyOf: - type: integer - type: 'null' title: End Page description: Ending page number token_count: anyOf: - type: integer - type: 'null' title: Token Count description: Number of tokens in chunk mime_type: anyOf: - type: string - type: 'null' title: Mime Type description: File MIME type category: anyOf: - type: string - type: 'null' title: Category description: File category tags: anyOf: - items: type: string type: array - type: 'null' title: Tags description: File tags document_type: anyOf: - type: string - type: 'null' title: Document Type description: Document type reference_id: anyOf: - type: string - type: 'null' title: Reference Id description: Client-provided file reference ID blob_storage_path: anyOf: - type: string - type: 'null' title: Blob Storage Path description: Full chunk JSON path in blob storage bounding_boxes: anyOf: - additionalProperties: true type: object - type: 'null' title: Bounding Boxes description: >- Bounding boxes for chunk highlighting: {content_id: {page: [x1, y1, x3, y3]}} type: object required: - chunk_id - file_id - file_name - content - score - chunk_index title: SearchResultChunk description: Individual search result chunk. StructuredResult: properties: columns: items: $ref: '#/components/schemas/ColumnInfo' type: array title: Columns description: Column metadata for the result set default: [] rows: items: additionalProperties: true type: object type: array title: Rows description: Result rows as key-value objects default: [] type: object title: StructuredResult description: Structured query results with schema metadata. ColumnInfo: properties: name: type: string title: Name description: Column name type: type: string title: Type description: Column data type (e.g., varchar, integer, timestamp) description: anyOf: - type: string - type: 'null' title: Description description: Human-readable column description type: object required: - name - type title: ColumnInfo description: Column metadata for structured results. securitySchemes: MemicApiKey: type: apiKey in: header name: X-API-Key description: >- Memic API key. Every key is bound to exactly one environment — get one from the Memic dashboard under API Keys. ````