Yongteng Lei 4cd4526492 Feat: PDF vision figure parser supports reading context (#12416) ...

### What problem does this PR solve?

PDF vision figure parser supports reading context.

### Type of change

- [x] New Feature (non-breaking change which adds functionality)

2026-01-05 09:55:43 +08:00

..

__init__.py

Fix IDE warnings (#12281)

2025-12-29 12:01:18 +08:00

analyze_task_system.md

Add license and Fix IDE warnings (#11985)

2025-12-17 17:04:44 +08:00

analyze_task_user.md

Feat: user defined prompt. (#9972)

2025-09-08 14:05:01 +08:00

ask_summary.md

Feat: Support metadata auto filer for Search. (#9524)

2025-08-19 10:27:24 +08:00

assign_toc_levels.md

Feat: TOC retrieval (#10456)

2025-10-10 17:07:55 +08:00

citation_plus.md

Feat: Redesign and refactor agent module (#9113)

2025-07-30 19:41:09 +08:00

citation_prompt.md

Fix typos (#11607)

2025-12-01 09:49:46 +08:00

content_tagging_prompt.md

Refa: refactor prompts into markdown-style structure using Jinja2 (#8667)

2025-07-04 15:59:41 +08:00

cross_languages_sys_prompt.md

Refa: refactor prompts into markdown-style structure using Jinja2 (#8667)

2025-07-04 15:59:41 +08:00

cross_languages_user_prompt.md

Refa: refactor prompts into markdown-style structure using Jinja2 (#8667)

2025-07-04 15:59:41 +08:00

full_question_prompt.md

Refa: refactor prompts into markdown-style structure using Jinja2 (#8667)

2025-07-04 15:59:41 +08:00

generator.py

Feat: PDF vision figure parser supports reading context (#12416)

2026-01-05 09:55:43 +08:00

keyword_prompt.md

Refa: refactor prompts into markdown-style structure using Jinja2 (#8667)

2025-07-04 15:59:41 +08:00

meta_data.md

fix(rag/prompts): Restructure metadata extraction rules for precision (#12360)

2025-12-31 13:52:33 +08:00

meta_filter.md

Feat: support operator in/not in for metadata filter. (#11503)

2025-11-25 12:44:26 +08:00

next_step.md

Fix:remove duplicate tool_meta (#12139)

2025-12-25 11:53:24 +08:00

question_prompt.md

Refa: refactor prompts into markdown-style structure using Jinja2 (#8667)

2025-07-04 15:59:41 +08:00

rank_memory.md

Feat: Redesign and refactor agent module (#9113)

2025-07-30 19:41:09 +08:00

reflect.md

Feat: user defined prompt. (#9972)

2025-09-08 14:05:01 +08:00

related_question.md

Feat: embedded search (#9501)

2025-08-18 12:05:11 +08:00

structured_output_prompt.md

Fix:output_structure in agent (#10907)

2025-11-03 09:39:53 +08:00

summary4memory.md

Feat: Redesign and refactor agent module (#9113)

2025-07-30 19:41:09 +08:00

template.py

Fix IDE warnings (#12281)

2025-12-29 12:01:18 +08:00

toc_detection.md

Feat: Use data pipeline to visualize the parsing configuration of the knowledge base (#10423)

2025-10-09 12:36:19 +08:00

toc_extraction_continue.md

Feat: Use data pipeline to visualize the parsing configuration of the knowledge base (#10423)

2025-10-09 12:36:19 +08:00

toc_extraction.md

Feat: Use data pipeline to visualize the parsing configuration of the knowledge base (#10423)

2025-10-09 12:36:19 +08:00

toc_from_text_system.md

Feat: TOC retrieval (#10456)

2025-10-10 17:07:55 +08:00

toc_from_text_user.md

Feat: Adds a new feature that enables the LLM to extract a structured table of contents (TOC) directly from plain text. (#10428)

2025-10-09 13:47:31 +08:00

toc_index.md

Feat: Use data pipeline to visualize the parsing configuration of the knowledge base (#10423)

2025-10-09 12:36:19 +08:00

toc_relevance_system.md

Feat: Add prompts for toc relevance check according to #10436 (#10457)

2025-10-10 11:44:46 +08:00

toc_relevance_user.md

Feat: Add prompts for toc relevance check according to #10436 (#10457)

2025-10-10 11:44:46 +08:00

tool_call_summary.md

Feat: Redesign and refactor agent module (#9113)

2025-07-30 19:41:09 +08:00

vision_llm_describe_prompt.md

Refa: refactor prompts into markdown-style structure using Jinja2 (#8667)

2025-07-04 15:59:41 +08:00

vision_llm_figure_describe_prompt_with_context.md

Feat: PDF vision figure parser supports reading context (#12416)

2026-01-05 09:55:43 +08:00

vision_llm_figure_describe_prompt.md

Feat: PDF vision figure parser supports reading context (#12416)

2026-01-05 09:55:43 +08:00