Commit graph

288 commits

Author SHA1 Message Date
Yiorgis Gozadinos
f4ddb0f5c6
Allow chat agent to specify the search limit 2026-01-12 15:52:40 +02:00
Yiorgis Gozadinos
2da2c6d13a
Search tool in chat agent needs no search context 2026-01-12 15:13:43 +02:00
Yiorgis Gozadinos
5adab04e50
Update prompts 2026-01-12 15:10:04 +02:00
Yiorgis Gozadinos
d43821659c
Filter low-confidence answers 2026-01-12 15:01:32 +02:00
Yiorgis Gozadinos
e4a7d86348
Rank q/a history with respect to cosine similarity to new queries. Adapt the chat agent to only pass relevant q/as 2026-01-12 14:15:36 +02:00
Yiorgis Gozadinos
7cc561d1db
Refactor to make agents a top-level module. Bring in the conversational agent from the app 2026-01-12 12:37:08 +02:00
Yiorgis Gozadinos
bb61e8fab6
Remove custom AG-UI infrastructure in favor of pydantic-ai native support 2026-01-12 12:37:07 +02:00
Yiorgis Gozadinos
26bc02ef14
Build a "conversational" research graph 2026-01-12 12:36:31 +02:00
Yiorgis Gozadinos
061095e855
Remove obsolete migrations 2026-01-12 12:07:49 +02:00
Yiorgis Gozadinos
29c682d8eb
Add migration 2026-01-11 11:40:57 +02:00
Yiorgis Gozadinos
9c4fd08c47
Update client serialization for compressed docling documents 2026-01-11 09:23:56 +02:00
Yiorgis Gozadinos
be715dd456
Revert docling core back to 1.8.0 for compatibility with docling-serve 2026-01-11 09:23:19 +02:00
Yiorgis Gozadinos
afc5b02285
Update schema to store docling_document as compressed bytes
- Change DocumentRecord.docling_document_json (str) to docling_document (bytes)
- Use large_binary Arrow type (64-bit offsets) to avoid 2GB column limit
- Decompress in Document.get_docling_document() using gzip
- Update DocumentRepository field mappings
2026-01-10 15:17:27 +02:00
Yiorgis Gozadinos
face6f95bc
compression utils 2026-01-10 15:11:58 +02:00
Yiorgis Gozadinos
09acc845c0
Update docling, docling-core, lancedb deps 2026-01-10 15:06:56 +02:00
Yiorgis Gozadinos
ff1f19bb08
vb 2026-01-08 18:50:39 +02:00
Yiorgis Gozadinos
3603b00513
Fix base64 images leaking into expanded search results
PictureItem.export_to_markdown() defaults to ImageRefMode.EMBEDDED,
which embeds base64 image data. When _extract_item_text() called this
method during context expansion, base64 data would leak into results.

Now explicitly handles PictureItem with ImageRefMode.PLACEHOLDER to
prevent base64 while still including VLM descriptions and captions.
2026-01-08 18:42:58 +02:00
Yiorgis Gozadinos
92f0fd956f
vb 2026-01-08 10:25:25 +02:00
Yiorgis Gozadinos
1616891c40
Fix non-reasoning OpenAI models that do not support reasoning config 2026-01-07 18:47:31 +02:00
Yiorgis Gozadinos
ef187767ed
vb 2026-01-07 11:20:48 +02:00
Yiorgis Gozadinos
2f413e937b
Add VLM model to download_models() 2026-01-07 11:09:24 +02:00
Yiorgis Gozadinos
f861664656
Extract shared DoclingServeClient for async workflow & use it for VLM support, chunking, converting 2026-01-07 11:09:23 +02:00
Yiorgis Gozadinos
769e0d4746
Update picture description prompt, move its config under prompts 2026-01-07 11:09:23 +02:00
Yiorgis Gozadinos
3c4116397d
Add VLM picture description support for image handling 2026-01-07 11:09:22 +02:00
Yiorgis Gozadinos
af340ac410
vb 2026-01-05 12:15:41 +02:00
Yiorgis Gozadinos
271af6cb3c
Update core dependencies 2026-01-05 11:52:38 +02:00
Yiorgis Gozadinos
5b34b9ed0e
Track potentially more than one running step in agui emitter 2026-01-05 10:41:35 +02:00
Yiorgis Gozadinos
679d0bfafd
Merge pull request #212 from ggozad/feat/contextualize-fts
Add additional context (headers/sections etc) to the FTS index
2025-12-29 18:42:19 +02:00
Yiorgis Gozadinos
a4d33c9e71
Delete test_agui_server.py, it was accidentally commited 2025-12-29 14:58:04 +02:00
Yiorgis Gozadinos
c9cb005d07
Use FTS on contextualized text 2025-12-29 14:50:07 +02:00
Yiorgis Gozadinos
1e9a10a235
Record cassettes for embedder, ignore huggingface while recording 2025-12-29 13:23:53 +02:00
Yiorgis Gozadinos
1f654917b8
vb 2025-12-26 13:56:29 +02:00
Yiorgis Gozadinos
08fb99c6f8
Add prompt customization and domain preable 2025-12-26 13:34:51 +02:00
Yiorgis Gozadinos
958fe43f2a
Remove vllm and lmstudio custom configs, they can now use the openai base 2025-12-26 12:02:55 +02:00
Yiorgis Gozadinos
86e31b8d29
VoyageAI embeddings, should end up as a PR for pydantic-ai 2025-12-26 11:50:47 +02:00
Yiorgis Gozadinos
3852e961b9
Update tests and client for explicit embed_query/embed_documents API 2025-12-26 11:26:56 +02:00
Yiorgis Gozadinos
132b8a36bc
Differentiate between embedding a query and a document 2025-12-24 12:43:11 +02:00
Yiorgis Gozadinos
5a30c197af
Delete obsolete embedders, rewrite get_embedder 2025-12-24 12:37:17 +02:00
Yiorgis Gozadinos
3141c91023
Add base_url to ModelConfig and EmbeddingModelConfig. Deprecate vllm and lm_studio configs, now through open ai 2025-12-24 12:33:44 +02:00
Yiorgis Gozadinos
271c50e225
pydantic-ai update 2025-12-24 12:25:22 +02:00
Yiorgis Gozadinos
5c7f033d2c
vb 2025-12-19 12:35:35 +02:00
Yiorgis Gozadinos
4164895514
Update critical dependencies 2025-12-19 12:21:57 +02:00
Yiorgis Gozadinos
2678c1aaad
history command, shows available lancedb versions per table 2025-12-19 12:03:18 +02:00
Yiorgis Gozadinos
77f9a2a1b9
before parameter in client, app, inspector, cli 2025-12-19 12:03:18 +02:00
Yiorgis Gozadinos
1e5eebfbf0
Checkout version logic and datetime utils 2025-12-19 12:03:16 +02:00
Yiorgis Gozadinos
8c0adb47e1
Merge branch 'main' into feat/read-only-db 2025-12-19 12:01:36 +02:00
Yiorgis Gozadinos
30f6bac388
Merge pull request #205 from ggozad/fix/docker-paths
Clarify how folder mounts work in docker docs
2025-12-19 11:59:00 +02:00
Yiorgis Gozadinos
090be293d2
Clarify folder mounts in docker docs, check monitor paths exist in FileWatcher 2025-12-19 11:58:08 +02:00
Yiorgis Gozadinos
1ff8fc1b56
Merge pull request #201 from ggozad/fix/allow-custom-config
Fix get_model() uses where default Config is forced
2025-12-19 11:57:30 +02:00
Yiorgis Gozadinos
fb1d3c97b3
Add additional programming language specifiers 2025-12-19 11:09:57 +02:00
Yiorgis Gozadinos
98727ad080
Add global --read-only mode in CLI 2025-12-18 16:43:49 +02:00
Yiorgis Gozadinos
1a8f72f99a
Add is_read_only property to Store. Guard all write operations in repositories 2025-12-18 15:55:24 +02:00
Yiorgis Gozadinos
e8e67d336c
Fix get_model() uses where default Config is forced 2025-12-18 15:10:31 +02:00
Yiorgis Gozadinos
1dbbaf6dd8
vb 2025-12-18 12:21:16 +02:00
Yiorgis Gozadinos
a1f5d944ba
Replace custom event classes with ag_ui.core types 2025-12-18 12:07:37 +02:00
Yiorgis Gozadinos
bf5711f7ab
Give hints in interactive cli 2025-12-18 11:48:35 +02:00
Yiorgis Gozadinos
08349e5c0e
QOL fixes for interactive cli research 2025-12-18 11:48:35 +02:00
Yiorgis Gozadinos
34092e4cdc
Wrap client side tool call in a message 2025-12-18 11:48:34 +02:00
Yiorgis Gozadinos
9818cac4eb
Fix agui tool call args being a json string 2025-12-18 11:48:34 +02:00
Yiorgis Gozadinos
218126de8d
Interactive research agent through AGUI client-side tool calls in CLI 2025-12-18 11:48:34 +02:00
Yiorgis Gozadinos
921893cff5
Remove deep ask from ag-ui server 2025-12-18 11:41:57 +02:00
Yiorgis Gozadinos
45de9bf0d5
Flatten graph structure, simplify 2025-12-18 11:41:57 +02:00
Yiorgis Gozadinos
557a296341
Update ag-ui example 2025-12-18 11:41:53 +02:00
Yiorgis Gozadinos
6e621ea34d
Provide full report on deep ask, modify prompt to not just summarize 2025-12-18 11:41:11 +02:00
Yiorgis Gozadinos
f0a7abd953
Remove deep qa graph, everything now uses the simplified research graph 2025-12-18 11:41:10 +02:00
Yiorgis Gozadinos
bdcf81774d
Remove analyze insights node, simplify research context, state, prompts & models 2025-12-18 11:41:10 +02:00
Yiorgis Gozadinos
50b7fb4461
Chunkers now set chunk.order directly 2025-12-18 11:35:48 +02:00
Yiorgis Gozadinos
3149552a61
Add 'plain' format and fallback for text conversion 2025-12-12 18:11:25 +02:00
Yiorgis Gozadinos
5c4799164c
Escape when getting by URI 2025-12-12 10:56:37 +02:00
Yiorgis Gozadinos
f6073a0db9
vb 2025-12-12 10:48:21 +02:00
Yiorgis Gozadinos
011f527579
Use rich rendering for citations in cli 2025-12-12 10:02:30 +02:00
Yiorgis Gozadinos
6f36085ac3
Fix synthesis prompt 2025-12-12 07:44:15 +02:00
Yiorgis Gozadinos
9ee8dfdc0c
Adapt prompts to clarify all lists must contain strings. Hopefully fixes LLM producing wrong output 2025-12-12 07:40:25 +02:00
Yiorgis Gozadinos
00d621e211
vb 2025-12-11 16:31:21 +02:00
Yiorgis Gozadinos
c0cbc22137
Info modal 2025-12-11 15:38:45 +02:00
Yiorgis Gozadinos
09be31f021
Lazy load chunks in inspector 2025-12-11 15:38:19 +02:00
Yiorgis Gozadinos
448d00f9b2
Add pagination to chunk repo's get_by_document_id 2025-12-11 15:38:19 +02:00
Yiorgis Gozadinos
c43c8fab22
Switch modals, do not stack them 2025-12-11 15:38:19 +02:00
Yiorgis Gozadinos
d7c755f454
Add filter to simple qa agent 2025-12-11 15:28:41 +02:00
Yiorgis Gozadinos
ecba8cfd38
Docs and cli support for filter in deep QA, Research 2025-12-11 15:23:50 +02:00
Yiorgis Gozadinos
5f4f5f70f0
Search filter in qa & research graphs 2025-12-11 14:37:43 +02:00
Yiorgis Gozadinos
7ccabf4fb9
Log to console with logfire only in development 2025-12-11 10:47:18 +02:00
Yiorgis Gozadinos
5edae13cc0
Run logfire regardless of environment 2025-12-11 10:43:25 +02:00
Yiorgis Gozadinos
2161727716
Add auto_vacuum config option to control automatic vacuuming 2025-12-10 11:42:14 +02:00
Yiorgis Gozadinos
ed809759d5
Fix mxbai rerank default model 2025-12-10 11:39:33 +02:00
Yiorgis Gozadinos
68457f251e
Do not generate base64 images by default 2025-12-09 18:42:56 +02:00
Yiorgis Gozadinos
156bba3359
Infinite scroll lazy load document list in inspector 2025-12-09 17:48:27 +02:00
Yiorgis Gozadinos
5a593e928e
Set max_context_items to 10 2025-12-09 12:47:42 +02:00
Yiorgis Gozadinos
802f058205
Move search related config under SearchConfig 2025-12-09 12:28:28 +02:00
Yiorgis Gozadinos
4da50873bd
Make search limit configurable 2025-12-09 11:46:43 +02:00
Yiorgis Gozadinos
a56e1ba67c
Simplify import_document, update_document 2025-12-08 18:19:12 +02:00
Yiorgis Gozadinos
67684c799b
Rebuild in batches to avoid creating multiple lancedb versions 2025-12-08 15:56:41 +02:00
Yiorgis Gozadinos
1b96937ec6
Update dependencies 2025-12-08 15:56:22 +02:00
Yiorgis Gozadinos
06b4f6a224
Add format parameter for text-to-DoclingDocument conversion 2025-12-08 15:56:21 +02:00
Yiorgis Gozadinos
1b14a4643d
Remove unused bounding box calculations 2025-12-08 15:56:21 +02:00
Yiorgis Gozadinos
9a5f8e4368
Add type-aware context expansion for search results 2025-12-08 15:56:21 +02:00
Yiorgis Gozadinos
dab04668bd
Dependency updates 2025-12-08 15:56:02 +02:00
Yiorgis Gozadinos
4f0f214f26
Remove unecessary tests, add no coverage pragmas where appropriate 2025-12-08 15:56:02 +02:00
Yiorgis Gozadinos
2572125804
Move embedding logic from ChunkRepository to client._ensure_chunks_embedded() 2025-12-08 15:56:02 +02:00
Yiorgis Gozadinos
e97d235c98
Unify update_document, update_document_fields 2025-12-08 15:56:02 +02:00