Yiorgis Gozadinos
969c45fe5c
Merge pull request #344 from ggozad/feat/docling-document-load-performace
...
Fast context expansion.
2026-04-16 12:33:52 +03:00
Yiorgis Gozadinos
75c84805b4
Improve judge prompt, QA prompt, and raise max_searches to 5
2026-04-16 12:12:23 +03:00
Yiorgis Gozadinos
e92f4774fa
Improve judge prompt
2026-04-16 12:11:54 +03:00
Yiorgis Gozadinos
d3c8ed62a8
preserve original chunk when expansion produces less content
2026-04-16 12:11:54 +03:00
Yiorgis Gozadinos
e2bac1e887
deduplicate escape_sql_string into utils
2026-04-16 12:11:53 +03:00
Yiorgis Gozadinos
39b9d0f683
Additional tests
2026-04-16 12:11:53 +03:00
Yiorgis Gozadinos
3c1a33779f
use windowed fetch for context expansion instead of loading all items
2026-04-16 12:11:53 +03:00
Yiorgis Gozadinos
b6113bf8ab
replace fixed-radius expansion with section-bounded algorithm
...
Context expansion is now automatic and structure-aware. For structured
documents, expands within the section containing the match. For sections
that exceed the budget or are too small, expands item-by-item outward
skipping noise labels. Unstructured documents use budget-based outward
expansion. Results sorted by relevance score.
2026-04-16 12:11:53 +03:00
Yiorgis Gozadinos
364b1bc509
add document_items table for fast context expansion
2026-04-16 12:11:53 +03:00
Yiorgis Gozadinos
661e0d34d8
Pin docling-core to avoid surprises with version bumps
2026-04-16 12:11:53 +03:00
Yiorgis Gozadinos
3bb10a5d22
Merge pull request #345 from ggozad/chore/bump-haiku-skills
...
Adapt to new haiku.skills
2026-04-16 12:07:18 +03:00
Yiorgis Gozadinos
e16c819e83
Adapt to new haiku.skills
2026-04-16 11:50:55 +03:00
Yiorgis Gozadinos
73d3807b7a
Merge pull request #343 from ggozad/chore/agui-update
...
Update ag-ui protocol
2026-04-14 14:00:42 +03:00
Yiorgis Gozadinos
0d6df472af
bump ag-ui protocol deps, fix duplicate React key in chat
2026-04-14 12:50:25 +03:00
Yiorgis Gozadinos
dcde86a5f1
vb
2026-04-09 15:33:19 +03:00
Yiorgis Gozadinos
0a4cd31168
Merge pull request #341 from ggozad/fix/skill-document-filtering
...
Fix: apply document filtering to list_documents & analyze tool
2026-04-09 15:30:51 +03:00
Yiorgis Gozadinos
a51b922be3
fix skill analyze tool ignores state.document_filter
2026-04-09 15:29:20 +03:00
Yiorgis Gozadinos
99af4fe11d
fix skill list_documents tool ignores state.document_filter
2026-04-09 15:29:20 +03:00
Yiorgis Gozadinos
4c1eec4ae7
Merge pull request #335 from ggozad/feat/s3-object-store
...
add S3/object storage support for LanceDB connections
2026-04-09 15:27:18 +03:00
Yiorgis Gozadinos
a118ee7aaa
Cleanup
2026-04-08 14:19:33 +03:00
Yiorgis Gozadinos
367accb814
Replace MinIO with SeaweedFS for S3 integration tests
2026-04-08 14:19:32 +03:00
Yiorgis Gozadinos
5e7a4ebae5
Update docs
2026-04-08 14:19:32 +03:00
Yiorgis Gozadinos
98641c24e4
S3 storage tests
2026-04-08 14:19:10 +03:00
Yiorgis Gozadinos
8e876ef28d
For remote stores, detect new db by checking if tables exist
2026-04-08 14:19:10 +03:00
Yiorgis Gozadinos
2522d46304
support remote storage in skill generator
2026-04-08 14:19:10 +03:00
Yiorgis Gozadinos
343bfd7199
guard app/UI filesystem checks for remote storage
2026-04-08 14:19:09 +03:00
Yiorgis Gozadinos
110accb8e7
Add ConnectionMode enum (LOCAL/CLOUD/OBJECT_STORAGE) and connect_lancedb() utility to support S3, GCS, Azure, and HDFS backends via storage_options.
2026-04-08 14:19:09 +03:00
Yiorgis Gozadinos
23d2aee955
Merge pull request #338 from ggozad/feat/zstd-image-separation
...
Separate page images into dedicated column and migrate to zstd compression
2026-04-08 14:16:12 +03:00
Yiorgis Gozadinos
d7d0090a92
cl
2026-04-08 14:08:00 +03:00
Yiorgis Gozadinos
900c10972b
Update docs
2026-04-08 14:04:33 +03:00
Yiorgis Gozadinos
3fe953eccb
Minor fixes
2026-04-08 14:04:32 +03:00
Yiorgis Gozadinos
9c5bef163e
Use zstandard get_frame_parameters for max_output_size
2026-04-08 14:04:32 +03:00
Yiorgis Gozadinos
62a436dd9d
vb
2026-04-08 14:04:32 +03:00
Yiorgis Gozadinos
89e2a303e6
migration to split pages and re-compress with zstd
2026-04-08 14:04:32 +03:00
Yiorgis Gozadinos
53445b8722
Add docling_pages column to DocumentRecord/Document for separate page image storage.
2026-04-08 14:04:32 +03:00
Yiorgis Gozadinos
b614ef19e1
switch compression from gzip to zstd, use stdlib for python >= 3.14
2026-04-08 14:04:32 +03:00
Yiorgis Gozadinos
f10d6b9960
Merge pull request #339 from ggozad/fix/generated-skills-preamble
...
fix: apply config.prompts.domain_preamble in generated skills
2026-04-08 14:03:24 +03:00
Yiorgis Gozadinos
f8db7fccb6
fix: apply config.prompts.domain_preamble in generated skills
2026-04-08 13:54:53 +03:00
Yiorgis Gozadinos
3d30840e9d
vb
2026-04-07 16:36:27 +03:00
Yiorgis Gozadinos
506fa050f6
Merge pull request #337 from ggozad/chore/search-performance
...
Performance improvements for large documents
2026-04-07 16:27:24 +03:00
Yiorgis Gozadinos
8a24606784
Strip page images from DoclingDocument before validation, unless we use visualize_chunk()
2026-04-07 12:54:23 +03:00
Yiorgis Gozadinos
32258d4e2a
Add batching to embeddings for huge documents
2026-04-07 12:06:28 +03:00
Yiorgis Gozadinos
68c3fa0f79
Add order to SearchResult, add ChunkRepository.get_chunks_in_range(), to use them in _expand_with_chunks to fetch only nearby chunks
2026-04-07 12:04:05 +03:00
Yiorgis Gozadinos
6fdb15e3b2
Add DocumentRepository.get_docling_data() for lazy loading docling in expand_context
2026-04-07 11:19:04 +03:00
Yiorgis Gozadinos
52712dbdb2
Use .select() projection to fetch only id/uri/title/metadata in search
2026-04-07 11:04:40 +03:00
Yiorgis Gozadinos
f2998ad0b5
add visualize_chunk test for multi-page chunks
2026-04-03 20:20:42 +03:00
Yiorgis Gozadinos
e286fc2053
Merge pull request #328 from ggozad/chore/deps-update
...
Update core dependencies
2026-04-03 11:44:29 +03:00
Yiorgis Gozadinos
e4dc60c4f1
Update core dependencies
2026-04-03 10:14:27 +03:00
Yiorgis Gozadinos
5c4164d224
vb
2026-04-01 13:32:56 +03:00
Yiorgis Gozadinos
71922c7cd4
Merge pull request #333 from ggozad/fix/skills-domain-preamble
...
Propagate domain_preamble to skill instructions and main agent preamble
2026-04-01 13:23:51 +03:00