Browsing Documents & Metadata
View Knowledge Base documents, system fields, and extracted metadata
In Knowledge Bases, select a Knowledge Base to browse its document list, search by name, and filter by status (pending/processing/completed/error); open a document to view details, chunks, and metadata.
Document System Fields
| Field | Description |
|---|---|
filename / name | Document name |
doc_type | File type: pdf, docx, doc, txt, markdown, html, csv, xlsx, xls, json, url |
file_size | File size in bytes |
status | pending, processing, completed, error |
chunk_count | Number of chunks generated |
token_count | Estimated token count |
source_url | Original URL (for documents added via URL) |
processed_at | When processing completed |
created_at / updated_at | Timestamps |
Extracted Metadata
The metadata JSON field is extracted automatically by the processing pipeline (e.g. title, author, language, section; the exact contents depend on the file format and extraction pipeline). Chunks also carry their own metadata (e.g. page number, section) used for retrieval context.
Editing
The only editable document field is the name:
curl -X PUT "https://your-domain.com/api/v1/knowledge-bases/$KB_ID/documents/$DOC_ID" \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{"name": "New name"}'The metadata JSON is regenerated during processing; there is no dedicated metadata editing endpoint.
Usage in Retrieval
Retrieval results and RAG context include each chunk's metadata. Search requests do not support metadata-based filtering or boosting — the only search filter is filter_doc_ids (restrict to specific document IDs).
See also:
- Import Documents — upload and processing
- Chunking & Processing — chunk preview and editing
- Retrieval & Quality Tuning — retrieval lab
How is this guide?