Complete outstanding ingestion and document UX work

README.md

@@ -111,10 +111,12 @@ Override via environment variables (`KB_ENGINE_URL`, `KB_API_KEY`) or CLI flags
 # Add notes
 kb addnote "Always restart nginx after config changes"
 kb addnote "Server room is building 3, floor 2" --tags ops
+kb addnote "Deploy checklist" --wait
 
-# Add files (async — uploads and exits immediately)
+# Add files (async by default; --wait blocks until ingestion finishes)
 kb addfile ~/docs/manual.pdf --tags admin
 kb addfile ~/notes/ --recursive
+kb addfile ~/docs/manual.pdf --wait
 
 # Check ingestion progress
 kb jobs
@@ -122,13 +124,16 @@ kb jobs
 # Search
 kb search "how to install git"
 kb search "deploy process" --tags ops --type pdf
+kb find "vehicle handbook" --type pdf
 
 # Update a note in place
 kb updatenote 42 "revised note content"
 
 # Manage
 kb list
-kb info 1
+kb list --title handbook
+kb list --filename M38T_PHEV
+kb info 1 --no-chunks
 kb tags
 kb tag 1 --add important
 kb export 1 -o manual.pdf    # download original file
@@ -160,6 +165,7 @@ The engine is configured via environment variables (set in the compose file or v
 | `KB_INGEST_DEVICE` | `auto` | Docling layout detection device: `auto`, `cpu`, or `cuda` |
 | `KB_API_KEY` | (none) | Optional Bearer token for API authentication |
 | `KB_SEARCH_THRESHOLD` | `0.01` | Minimum score for search results (filters noise) |
+| `KB_MIN_CHUNK_ALNUM` | `3` | Minimum alphanumeric characters retained in Docling PDF/DOCX/HTML chunks (`0` disables) |
 | `KB_RERANK_ENABLED` | `false` (`true` in nvidia compose) | Load a cross-encoder and rerank hybrid results server-side |
 | `KB_RERANKER_MODEL` | `BAAI/bge-reranker-v2-m3` | Cross-encoder model for reranking |
 | `KB_RERANK_CANDIDATES` | `40` | Hybrid candidates scored by the reranker per query |
@@ -169,6 +175,21 @@ The engine is configured via environment variables (set in the compose file or v
 | `HF_HUB_OFFLINE` | (none) | Set to `1` to prevent model downloads (use cached only) |
 | `KB_DATA_PATH` | `./data` | Host path for bind mount (compose variable, not used by engine) |
 
+### Repairing legacy note titles
+
+Notes created by older clients may have a synthetic `<uuid>_note.note` title.
+The maintenance command previews only unambiguous matches by default:
+
+```bash
+cd engine
+python -m kb.maintenance.backfill_note_titles
+python -m kb.maintenance.backfill_note_titles --apply
+```
+
+The apply mode reloads the configured embedding model and refreshes each
+affected note's title, enriched full-text content, and vector embedding. Back
+up the data directory before running it against production.
+
 ## Data portability
 
 The data directory contains everything: SQLite database, model cache, and staging files. To migrate between hosts: