用于构建垂直领域知识库的逻辑推理问答解决方案KAG is a logical form-guided reasoning and retrieval framework based on OpenSPG engine and LLMs. It is used to build logical reasoning and factual Q&A solutions for professional domain knowledge bases. It can effectively overcome the shortcomings of the traditional RAG vector similarity calculation model.
active 2024-09-21 → 2026-08-13 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 692 days from 2024-09-21 to 2026-08-13. Pushes: 1,191 total, peak 26 in a day. Pull requests: 449 total, peak 26 in a day. Issues: 619 total, peak 45 in a day. Comments: 974 total, peak 34 in a day. Stars: 7,587 total, peak 584 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| caszkgui | 711 | 164 | 89 | 208 |
| zhuzhongshu123 | 446 | 223 | 176 | 12 |
| royzhao | 442 | 313 | 63 | 15 |
| northmachine | 316 | 234 | 45 | 9 |
| xionghuaidong | 218 | 102 | 30 | 46 |
| andylau-55 | 115 | 27 | 10 | 24 |
| thundax-lyp | 99 | 0 | 5 | 93 |
| wxy20152016 | 56 | 52 | 1 | 0 |
| BQQQQB | 37 | 0 | 0 | 15 |
| jerryHo123 | 35 | 0 | 0 | 18 |
| 5478-lfy | 30 | 0 | 0 | 16 |
| github-actions[bot] | 30 | 0 | 0 | 0 |
| ikkvix | 26 | 26 | 0 | 0 |
| xiaozhou123-oos | 17 | 0 | 0 | 13 |
| fishfish123-win | 17 | 0 | 0 | 13 |
| tpoisonooo | 15 | 0 | 3 | 9 |
| werruww | 14 | 0 | 0 | 13 |
| thistleknot | 14 | 0 | 0 | 9 |
| yuuki-nanamin | 14 | 0 | 0 | 6 |
| Gotodie | 13 | 0 | 0 | 12 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#740qhonge2026-05-07 09:10这个项目还在维护吗
- Pull request#745octo-patch2026-04-20 03:08
- Pull request#741octo-patch2026-03-20 02:02
- Issue#740Vickzhang2026-02-09 07:12这个项目还在维护吗
- Issue comment#694Kaedeser2026-01-31 10:33结构化任务
- Issue comment#737Kaedeser2026-01-31 10:13抽取的实体名写入图数据库时，id和name过滤掉了标点符号，希望保留这些标点符号话，应该在哪里修改代码
- Pull request#736Codingwqy2026-01-28 13:30
- Issue comment#634wstars19942026-01-22 09:12开发者模式新建知识库，再产品模型应用里，用这个知识库没效果
- Issue#7374025007332026-01-19 07:45抽取的实体名写入图数据库时，id和name过滤掉了标点符号，希望保留这些标点符号话，应该在哪里修改代码
- Issue comment#703Kaedeser2026-01-12 12:45如何根据前端页面中显示的抽取效果去提取实体之间的关系？
- Issue comment#726Codingwqy2026-01-10 01:50[Bug] [Module Name] Bug title 用户手册快速开始中的超链接全部 404
- Issue comment#730Codingwqy2026-01-10 01:22创建知识库性能慢
- Issue comment#719Kaedeser2026-01-09 14:12[Bug] [Module Name] Bug title 开发者模式下，知识库建立出错
- Issue comment#719Kaedeser2026-01-09 14:10[Bug] [Module Name] Bug title 开发者模式下，知识库建立出错
- Issue comment#728Kaedeser2026-01-07 14:34No Graph in the Knowledge Exploration section
- Issue comment#473Codingwqy2026-01-06 07:25如何处理结构化数据的关联,如多表关联
- Issue comment#490Codingwqy2026-01-06 07:07问答api接口：深度推理和简单推理如何传参
- Issue#734a-big-cup2025-12-30 06:43[Bug] [Module Name] Bug title 'NoneType' object is not iterable
- Issue comment#734a-big-cup2025-12-30 06:43[Bug] [Module Name] Bug title 'NoneType' object is not iterable
- Issue comment#733macaulishxcoo2025-12-29 08:01[Bug] [Module Name] Bug title knext.common.rest.exceptions.ApiException: (400)
- Issue#733macaulishxcoo2025-12-29 07:50[Bug] [Module Name] Bug title knext.common.rest.exceptions.ApiException: (400)
- Issue comment#731dms03242025-12-26 03:52[version0.8] 2wiki创建project报错
- Issue#7317748587122025-12-26 02:58[version0.8] 2wiki创建project报错
- Issue comment#714zhangwh8072025-12-19 03:19知识抽取出现问题了，怎么解决
- Issue comment#627RichardASimon2025-12-18 07:09[Bug] [Module Name] 在前端配置嵌入模型时报错
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 7,587 stars here means stars gained during the window, not the repo's star count.