> ## Documentation Index > Fetch the complete documentation index at: https://docs.noxus.ai/llms.txt > Use this file to discover all available pages before exploring further. # Get Knowledge Bases V2 ## OpenAPI ````yaml get /v1/knowledge-bases openapi: 3.1.0 info: title: Noxus Backend description: >- Backend API for the Noxus Platform. More information can be found at https://docs.noxus.ai version: 1.1.0 servers: [] security: [] paths: /v1/knowledge-bases: get: tags: - V1 - Knowledge Bases summary: Get Knowledge Bases V2 operationId: get_knowledge_bases_v2 parameters: - name: query in: query required: false schema: anyOf: - type: string - type: 'null' title: Query - name: order_by in: query required: false schema: type: string default: last_updated_at title: Order By - name: sort in: query required: false schema: type: string default: desc title: Sort - name: page in: query required: false schema: type: integer minimum: 1 description: Page number default: 1 title: Page description: Page number - name: size in: query required: false schema: type: integer maximum: 100 minimum: 1 description: Page size default: 50 title: Size description: Page size responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/Page_KnowledgeBaseV2_' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - APIKeyHeader: [] components: schemas: Page_KnowledgeBaseV2_: properties: items: items: $ref: '#/components/schemas/KnowledgeBaseV2' type: array title: Items total: type: integer minimum: 0 title: Total page: type: integer minimum: 1 title: Page size: type: integer minimum: 1 title: Size pages: type: integer minimum: 0 title: Pages type: object required: - items - total - page - size - pages title: Page[KnowledgeBaseV2] HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError KnowledgeBaseV2: properties: id: type: string format: uuid title: Id group_id: type: string format: uuid title: Group Id name: type: string title: Name status: $ref: '#/components/schemas/TrainStatus' description: type: string title: Description document_types: items: type: string type: array title: Document Types kb_type: $ref: '#/components/schemas/KBType' size: type: number title: Size num_docs: type: integer title: Num Docs created_at: type: string format: date-time title: Created At updated_at: type: string format: date-time title: Updated At error: anyOf: - additionalProperties: true type: object - type: 'null' title: Error version: type: string enum: - v2 - v3 title: Version meta: anyOf: - additionalProperties: true type: object - type: 'null' title: Meta embedding_health: anyOf: - $ref: '#/components/schemas/EmbeddingHealth' - type: 'null' health: anyOf: - $ref: '#/components/schemas/KBHealth' - type: 'null' total_documents: type: integer title: Total Documents training_documents: type: integer title: Training Documents trained_documents: type: integer title: Trained Documents error_documents: type: integer title: Error Documents uploaded_documents: type: integer title: Uploaded Documents source_types: additionalProperties: type: integer type: object title: Source Types training_source_types: items: type: string type: array title: Training Source Types settings_: anyOf: - $ref: '#/components/schemas/KBConfig' - type: 'null' type: object required: - id - group_id - name - status - description - document_types - kb_type - size - num_docs - created_at - updated_at - error - version - total_documents - training_documents - trained_documents - error_documents - uploaded_documents - source_types - training_source_types title: KnowledgeBaseV2 ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError TrainStatus: type: string enum: - created - trained - training - error title: TrainStatus KBType: type: string enum: - temporary - entity title: KBType EmbeddingHealth: properties: healthy: type: boolean title: Healthy message: type: string title: Message default: '' embedding_model: type: string title: Embedding Model default: '' type: object required: - healthy title: EmbeddingHealth KBHealth: properties: status: type: string enum: - ok - needs_repair - error title: Status issue: anyOf: - type: string enum: - vector_table_missing - index_struct_missing - dim_drift - resolver_error - embedding_config_error - type: 'null' title: Issue detected_dim: anyOf: - type: integer - type: 'null' title: Detected Dim declared_dim: anyOf: - type: integer - type: 'null' title: Declared Dim message: type: string title: Message default: '' action: type: string title: Action default: '' type: object required: - status title: KBHealth description: |- Structural health of a v3 knowledge base, computed at request time. Returned on every ``GET /kbs/{id}`` so the FE can surface an inline banner + CTA when the KB needs repair. Four concrete issues can be detected by ``KBManagementService.compute_health``: - ``vector_table_missing`` — pgvector table was never created or was dropped out from under us. Repair rebuilds the table at the declared dim and queues every document for re-ingestion. - ``index_struct_missing`` — the pgvector table is present but the kirby ``index_struct`` row is absent. Repair rebuilds the struct anchored to the existing column; no documents need reprocessing. - ``dim_drift`` — declared dim no longer matches the pgvector column width. Retrieval still works against the actual dim, but new INSERTs fail. Repair recreates the table at the declared dim and queues every document for re-ingestion. - ``resolver_error`` — the resolver itself failed (DB outage, unexpected column type). Not a repair-able state; the user is asked to retry later. HTTP stays 200 so the FE can render the apology inline instead of a blank error page. KBConfig: properties: retrieval_info: type: string title: Retrieval Info description: Information about retrieval settings default: '' display: col_span: 12 is_disabled: false is_subconfig: false label: '' needs_confirmation: false show_optional_label: false small_text: >- Configure how documents are retrieved from your Knowledge Base. These settings set the defaults when using a specific retrieval method. title: Retrieval settings type: ui_text_display tab: Retrieval tool_mode: hidden retrieval_method: type: string title: Retrieval Method default: hybrid_bm25 display: col_span: 12 icons: false is_clearable: false is_disabled: false is_horizontal: false is_searchable: true is_subconfig: false label: Retrieval method needs_confirmation: false show_optional_label: false type: select values: - description: Uses vector embeddings for meaning-based search label: Semantic search value: semantic - description: Traditional keyword-based search with ranking label: Keyword search value: full_text_search - description: Keyword search ranked with BM25 (IDF + term saturation) label: BM25 keyword search value: bm25 - description: Combines semantic and keyword search with score fusion label: Hybrid search value: hybrid_relative_score - description: Combines semantic and keyword search with RRF score fusion label: RRF hybrid search value: hybrid_rrf - description: Combines semantic and BM25 keyword search with score fusion label: Hybrid search (BM25) value: hybrid_bm25 - description: Combines semantic and BM25 keyword search with RRF fusion label: RRF hybrid search (BM25) value: hybrid_bm25_rrf rules: - type: sets_connector_or_config_visibility tab: Retrieval tool_mode: hidden reranking_enabled: type: boolean title: Reranking Enabled description: >- Retrieve a wider candidate pool, then reorder it with a reranker before returning results. default: false display: col_span: 12 is_disabled: false is_subconfig: false label: Enable reranking needs_confirmation: false show_optional_label: false toggle_position: right type: toggle rules: - type: sets_connector_or_config_visibility tab: Retrieval tool_mode: hidden rerankers: items: type: string type: array title: Rerankers description: >- Tried in order — if one is unavailable the next takes over. Leave empty to use the rerankers enabled for your workspace. display: col_span: 12 endpoint: /models/rerankers hide_disabled_models: true is_disabled: false is_subconfig: false label: Rerankers layout: side_by_side model_limit: 3 model_type: rerankers needs_confirmation: false show_inline_presets: false show_optional_label: false show_presets: false type: model_select rules: - config_source: reranking_enabled type: visible_if value: true tab: Retrieval tool_mode: hidden retrieval_top_k: type: integer title: Retrieval Top K description: Maximum number of results to return default: 25 display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Max results max: 100 min: 1 needs_confirmation: false show_optional_label: false step: 1 type: number_slider tab: Retrieval tool_mode: hidden retrieval_top_k_window: items: type: integer type: array title: Retrieval Top K Window description: >- Searches against this Knowledge Base always return between min and max results — requested top_k values outside this window are clamped. display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Allowed results range max: 100 min: 1 needs_confirmation: false show_optional_label: false step: 1 type: number_range_slider tab: Retrieval tool_mode: hidden fts_weight: type: number title: Fts Weight description: Weight for keyword search in hybrid methods default: 0.7 display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Keyword search weight max: 1 min: 0 needs_confirmation: false show_optional_label: false step: 0.1 type: number_slider rules: - conds: - config_source: retrieval_method type: visible_if value: hybrid_relative_score - config_source: retrieval_method type: visible_if value: hybrid_reranking - config_source: retrieval_method type: visible_if value: hybrid_bm25 mode: or type: visible_if_group tab: Retrieval tool_mode: hidden semantic_weight: type: number title: Semantic Weight description: Weight for semantic search in hybrid methods default: 0.3 display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Semantic search weight max: 1 min: 0 needs_confirmation: false show_optional_label: false step: 0.1 type: number_slider rules: - conds: - config_source: retrieval_method type: visible_if value: hybrid_relative_score - config_source: retrieval_method type: visible_if value: hybrid_reranking - config_source: retrieval_method type: visible_if value: hybrid_bm25 mode: or type: visible_if_group tab: Retrieval tool_mode: hidden rrf_k: type: integer title: Rrf K description: RRF constant k - higher values reduce the impact of rank differences default: 60 display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: RRF constant (k) max: 200 min: 1 needs_confirmation: false show_optional_label: false step: 1 type: number_slider rules: - conds: - config_source: retrieval_method type: visible_if value: hybrid_rrf - config_source: retrieval_method type: visible_if value: hybrid_bm25_rrf mode: or type: visible_if_group tab: Retrieval tool_mode: hidden threshold_enabled: type: boolean title: Threshold Enabled description: >- The system will only return results with a score above the defined threshold. default: false display: col_span: 12 is_disabled: false is_subconfig: false label: Enable score threshold needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Retrieval tool_mode: hidden visible: false threshold: type: number title: Threshold description: Minimum score threshold for results default: 0 display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Score threshold max: 1 min: 0 needs_confirmation: false show_optional_label: false step: 0.01 type: number_slider rules: - config_source: threshold_enabled type: visible_if value: true tab: Retrieval tool_mode: hidden reranker_type: type: string title: Reranker Type description: The reranking method to use after initial retrieval default: vertex display: col_span: 12 icons: false is_clearable: false is_disabled: false is_horizontal: false is_searchable: true is_subconfig: false label: Reranker type needs_confirmation: false show_optional_label: false type: select values: - description: Google's semantic ranking service label: Vertex AI reranker value: vertex rules: - config_source: retrieval_method type: visible_if value: hybrid_reranking - type: sets_connector_or_config_visibility tab: Retrieval tool_mode: hidden visible: false reranker_top_k: type: integer title: Reranker Top K description: Number of results to return from reranker default: 10 display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Reranker top k max: 25 min: 1 needs_confirmation: false show_optional_label: false step: 1 type: number_slider rules: - config_source: retrieval_method type: visible_if value: hybrid_reranking tab: Retrieval tool_mode: hidden visible: false reranker_threshold_enabled: type: boolean title: Reranker Threshold Enabled description: Enable minimum score threshold for reranker default: false display: col_span: 12 is_disabled: false is_subconfig: false label: Enable reranker score threshold needs_confirmation: false show_optional_label: false toggle_position: right type: toggle rules: - config_source: retrieval_method type: visible_if value: hybrid_reranking tab: Retrieval tool_mode: hidden visible: false reranker_threshold: type: number title: Reranker Threshold description: Minimum score threshold for reranker results default: 0 display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Reranker score threshold max: 1 min: 0 needs_confirmation: false show_optional_label: false step: 0.01 type: number_slider rules: - conds: - config_source: retrieval_method type: visible_if value: hybrid_reranking - config_source: reranker_threshold_enabled type: visible_if value: true mode: and type: visible_if_group tab: Retrieval tool_mode: hidden visible: false vertex_model: type: string title: Vertex Model description: Vertex AI ranking model to use default: semantic-ranker-512@latest display: col_span: 12 icons: false is_clearable: false is_disabled: false is_horizontal: false is_searchable: true is_subconfig: false label: Vertex model needs_confirmation: false show_optional_label: false type: select values: - label: Semantic ranker 512 value: semantic-ranker-512@latest rules: - conds: - config_source: reranker_type type: visible_if value: vertex - config_source: retrieval_method type: visible_if value: hybrid_reranking mode: and type: visible_if_group tab: Retrieval tool_mode: hidden visible: false ingestion_info: type: string title: Ingestion Info description: Information about ingestion settings default: '' display: col_span: 12 is_disabled: false is_subconfig: false label: '' needs_confirmation: false show_optional_label: false small_text: >- Configure how documents are processed and stored in your Knowledge Base. These settings control chunking, enrichment, and file-specific processing options. title: Ingestion settings type: ui_text_display optional: true tab: Ingestion tool_mode: hidden batch_size: type: integer title: Batch Size description: Number of documents to process in each batch default: 8 display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Batch size max: 32 min: 1 needs_confirmation: false show_optional_label: false step: 1 type: number_slider tab: Ingestion tool_mode: hidden visible: false default_chunk_size: type: integer title: Default Chunk Size description: Default size for text chunks in characters default: 2048 display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Default chunk size max: 4096 min: 256 needs_confirmation: false show_optional_label: false step: 256 type: number_slider tab: Ingestion tool_mode: hidden default_chunk_overlap: type: integer title: Default Chunk Overlap description: Default overlap between chunks in characters default: 512 display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Default chunk overlap max: 1024 min: 0 needs_confirmation: false show_optional_label: false step: 64 type: number_slider tab: Ingestion tool_mode: hidden enrich_chunks_mode: type: string title: Enrich Chunks Mode description: How to enrich chunks with additional context default: inject_summary display: col_span: 12 icons: false is_clearable: false is_disabled: false is_horizontal: false is_searchable: true is_subconfig: false label: Chunk enrichment mode needs_confirmation: false show_optional_label: false type: select values: - description: Add document summary to each chunk for better context label: Inject summary value: inject_summary - description: Generate contextual information for each chunk label: Contextual value: contextual - description: >- Skip per-document summary generation entirely. Cuts ingestion latency and LLM cost dramatically for small documents at the cost of slightly worse retrieval on large documents. label: None value: none tab: Ingestion tool_mode: hidden visible: false enrich_pre_made_qa: type: boolean title: Enrich Pre Made Qa description: Generate Q&A pairs from the document content default: false display: col_span: 12 is_disabled: false is_subconfig: false label: Generate Q&A pairs needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false pdf_info: type: string title: Pdf Info description: PDF-specific processing settings default: '' accordion: PDF Options display: col_span: 12 is_disabled: false is_subconfig: false label: '' needs_confirmation: false show_optional_label: false small_text: >- Configure settings specific to PDF document processing including OCR, image extraction, and chunking parameters. title: PDF Processing type: ui_text_display optional: true tab: Ingestion tool_mode: hidden visible: false pdf_extract_images: type: boolean title: Pdf Extract Images description: Extract and process images from PDF files default: true accordion: PDF Options display: col_span: 12 is_disabled: false is_subconfig: false label: Extract images from PDFs needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false pdf_chunk_size: type: integer title: Pdf Chunk Size description: Chunk size for PDF documents default: 1024 accordion: PDF Options display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: PDF chunk size max: 4096 min: 256 needs_confirmation: false show_optional_label: false step: 256 type: number_slider tab: Ingestion tool_mode: hidden visible: false pdf_chunk_overlap: type: integer title: Pdf Chunk Overlap description: Chunk overlap for PDF documents default: 256 accordion: PDF Options display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: PDF chunk overlap max: 1024 min: 0 needs_confirmation: false show_optional_label: false step: 64 type: number_slider tab: Ingestion tool_mode: hidden visible: false pdf_ocr_enabled: type: boolean title: Pdf Ocr Enabled description: Enable OCR processing for scanned PDFs default: false accordion: PDF Options display: col_span: 12 is_disabled: false is_subconfig: false label: Enable OCR needs_confirmation: false show_optional_label: false toggle_position: right type: toggle rules: - type: sets_connector_or_config_visibility tab: Ingestion tool_mode: hidden visible: false pdf_ocr_confidence: type: number title: Pdf Ocr Confidence description: Minimum confidence threshold for OCR text recognition default: 0.7 accordion: PDF Options display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: OCR confidence threshold max: 1 min: 0.1 needs_confirmation: false show_optional_label: false step: 0.1 type: number_slider rules: - config_source: pdf_ocr_enabled type: visible_if value: true tab: Ingestion tool_mode: hidden visible: false pdf_ocr_language: type: string title: Pdf Ocr Language description: Language hint for OCR processing default: auto accordion: PDF Options display: col_span: 12 icons: false is_clearable: false is_disabled: false is_horizontal: false is_searchable: true is_subconfig: false label: OCR language needs_confirmation: false show_optional_label: false type: select values: - label: Auto-detect value: auto - label: English value: en - label: Spanish value: es - label: French value: fr - label: German value: de - label: Portuguese value: pt - label: Chinese value: zh - label: Japanese value: ja rules: - config_source: pdf_ocr_enabled type: visible_if value: true tab: Ingestion tool_mode: hidden visible: false text_info: type: string title: Text Info description: Text-specific processing settings default: '' accordion: Plain Text Options display: col_span: 12 is_disabled: false is_subconfig: false label: '' needs_confirmation: false show_optional_label: false small_text: >- Configure settings for plain text document processing including chunking and formatting options. title: Text Processing type: ui_text_display optional: true tab: Ingestion tool_mode: hidden visible: false text_chunking_enabled: type: boolean title: Text Chunking Enabled description: Enable automatic chunking for text files default: true accordion: Plain Text Options display: col_span: 12 is_disabled: false is_subconfig: false label: Enable text chunking needs_confirmation: false show_optional_label: false toggle_position: right type: toggle rules: - type: sets_connector_or_config_visibility tab: Ingestion tool_mode: hidden visible: false text_chunk_size: type: integer title: Text Chunk Size description: Chunk size for text documents default: 1024 accordion: Plain Text Options display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Text chunk size max: 4096 min: 256 needs_confirmation: false show_optional_label: false step: 256 type: number_slider rules: - config_source: text_chunking_enabled type: visible_if value: true tab: Ingestion tool_mode: hidden visible: false text_chunk_overlap: type: integer title: Text Chunk Overlap description: Chunk overlap for text documents default: 256 accordion: Plain Text Options display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Text chunk overlap max: 1024 min: 0 needs_confirmation: false show_optional_label: false step: 64 type: number_slider rules: - config_source: text_chunking_enabled type: visible_if value: true tab: Ingestion tool_mode: hidden visible: false text_preserve_formatting: type: boolean title: Text Preserve Formatting description: Preserve original text formatting and structure default: true accordion: Plain Text Options display: col_span: 12 is_disabled: false is_subconfig: false label: Preserve formatting needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false html_info: type: string title: Html Info description: HTML-specific processing settings default: '' accordion: HTML ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: '' needs_confirmation: false show_optional_label: false small_text: >- Configure settings for HTML document processing including content extraction and chunking parameters. title: HTML Processing type: ui_text_display optional: true tab: Ingestion tool_mode: hidden visible: false html_extract_text_only: type: boolean title: Html Extract Text Only description: Extract only text content from HTML default: true accordion: HTML ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: Extract text only needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false html_include_tables: type: boolean title: Html Include Tables description: Include table content from HTML default: true accordion: HTML ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: Include tables needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false html_include_images: type: boolean title: Html Include Images description: Include image references from HTML default: false accordion: HTML ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: Include images needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false html_ingest_as_markdown: type: boolean title: Html Ingest As Markdown description: >- Convert HTML to Markdown before chunking. Preserves headings, lists, links, and tables while dropping tag noise — produces smaller, more semantic chunks. Recommended for crawled web pages; turn off only to reproduce legacy raw-HTML behaviour. default: true accordion: HTML ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: Ingest as Markdown needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden image_info: type: string title: Image Info description: Image-specific processing settings default: '' accordion: Image Options display: col_span: 12 is_disabled: false is_subconfig: false label: '' needs_confirmation: false show_optional_label: false small_text: >- Configure settings for image document processing including vision model settings and content extraction. title: Image processing type: ui_text_display optional: true tab: Ingestion tool_mode: hidden visible: false image_ocr_enabled: type: boolean title: Image Ocr Enabled description: Enable OCR processing for images default: false accordion: Image Options display: col_span: 12 is_disabled: false is_subconfig: false label: Enable OCR for images needs_confirmation: false show_optional_label: false toggle_position: right type: toggle rules: - type: sets_connector_or_config_visibility tab: Ingestion tool_mode: hidden visible: false image_ocr_confidence: type: number title: Image Ocr Confidence description: Minimum confidence threshold for image OCR default: 0.7 accordion: Image Options display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Image OCR confidence max: 1 min: 0.1 needs_confirmation: false show_optional_label: false step: 0.1 type: number_slider rules: - config_source: image_ocr_enabled type: visible_if value: true tab: Ingestion tool_mode: hidden visible: false csv_row_as_document: type: boolean title: Csv Row As Document description: Treat each row as a separate chunk. default: false accordion: CSV ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: Row as chunk needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false csv_include_headers: type: boolean title: Csv Include Headers description: Include headers in the text representation default: true accordion: CSV ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: Include headers needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false csv_auto_detect_delimiter: type: boolean title: Csv Auto Detect Delimiter description: Automatically detect CSV delimiter default: true accordion: CSV ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: Auto-detect delimiter needs_confirmation: false show_optional_label: false toggle_position: right type: toggle rules: - type: sets_connector_or_config_visibility tab: Ingestion tool_mode: hidden visible: false csv_delimiter: type: string title: Csv Delimiter description: Custom delimiter for CSV parsing default: ',' accordion: CSV ingestion options display: col_span: 12 icons: false is_clearable: false is_disabled: false is_horizontal: false is_searchable: true is_subconfig: false label: CSV delimiter needs_confirmation: false show_optional_label: false type: select values: - label: Comma (,) value: ',' - label: Semicolon (;) value: ; - label: Tab value: "\t" - label: Pipe (|) value: '|' rules: - config_source: csv_auto_detect_delimiter type: visible_if value: false tab: Ingestion tool_mode: hidden visible: false csv_encoding: type: string title: Csv Encoding description: File encoding for CSV files default: utf-8 accordion: CSV ingestion options display: col_span: 12 icons: false is_clearable: false is_disabled: false is_horizontal: false is_searchable: true is_subconfig: false label: CSV encoding needs_confirmation: false show_optional_label: false type: select values: - label: UTF-8 value: utf-8 - label: ASCII value: ascii - label: ISO-8859-1 value: iso-8859-1 - label: Windows-1252 value: cp1252 tab: Ingestion tool_mode: hidden visible: false xlsx_row_as_document: type: boolean title: Xlsx Row As Document description: >- Excel: emit one chunk per row (with cell coordinates and detected headers). Off → render the whole sheet as a markdown table and chunk by token budget — closer to the legacy text-based behavior. Default on; toggle off if a workbook has unusual structure that confuses row mode. default: false accordion: Excel ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: Row as chunk needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false xlsx_include_headers: type: boolean title: Xlsx Include Headers description: >- Excel: include detected column headers alongside cell coordinates in row chunks. default: true accordion: Excel ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: Include headers needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false tabular_ingest_all_rows: type: boolean title: Tabular Ingest All Rows description: >- Ingest every row regardless of size. Turn on only for small-to-medium workbooks; very large files will slow retrieval and may exhaust worker memory. default: false accordion: Tabular ingestion options display: col_span: 12 is_disabled: false is_subconfig: false label: Ingest all rows needs_confirmation: false show_optional_label: false toggle_position: right type: toggle tab: Ingestion tool_mode: hidden visible: false tabular_max_rows: type: integer title: Tabular Max Rows description: >- Excel & CSV: max rows ingested per sheet when the 'Ingest all rows' toggle is off. Rows beyond this are dropped and the document is flagged as truncated. default: 50000 accordion: Tabular ingestion options display: col_span: 12 hide_number: true is_disabled: false is_subconfig: false label: Max rows per sheet max: 500000 min: 100 needs_confirmation: false show_optional_label: false step: 100 type: number_slider rules: - config_source: tabular_ingest_all_rows type: visible_if value: false tab: Ingestion tool_mode: hidden visible: false embedding_model: items: type: string type: array title: Embedding Model description: The model to use for embedding documents default: [] display: col_span: 12 endpoint: /models/embeddings hide_disabled_models: false is_disabled: false is_subconfig: false label: Embedding model layout: side_by_side model_limit: 1 model_type: embeddings needs_confirmation: false show_inline_presets: false show_optional_label: false show_presets: false type: model_select tab: Ingestion tool_mode: hidden embedding_dimension: anyOf: - type: integer - type: 'null' title: Embedding Dimension description: Leave unset to use the model's default dimension. display: col_span: 12 endpoint: /models/embeddings fallback_values: [] is_clearable: false is_disabled: false is_searchable: false is_subconfig: false label: Embedding dimension needs_confirmation: false placeholder: Use model default show_optional_label: false source_attribute: supported_dimensions source_field: embedding_model type: config_dynamic_select optional: true tab: Ingestion tool_mode: hidden mm_embedding_model: items: type: string type: array title: Mm Embedding Model description: The model to use for embedding documents default: [] display: col_span: 12 endpoint: /models/embeddings?multimodal=true hide_disabled_models: true is_disabled: false is_subconfig: false label: Embedding model layout: side_by_side model_limit: 3 model_type: embeddings needs_confirmation: false show_inline_presets: false show_optional_label: false show_presets: false type: model_select tab: Ingestion tool_mode: hidden visible: false type: object title: KBConfig securitySchemes: APIKeyHeader: type: apiKey in: header name: X-API-Key ````