clearai-dsh 0.2.0 → 0.2.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md CHANGED
@@ -2,9 +2,36 @@
2
2
 
3
3
  All notable changes to this project are recorded here. The format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
4
4
 
5
- # Changelog
5
+ ## [0.2.1] — 2026-09-18
6
+
7
+ **知识任务是循环的原生行为,不是另一个模式。** 本体、实体、认识论早就在,但普通研究的最短路径仍然是「检索 → 总结 → 写报告」——要建本体得用户先想起来说一句。这一版修的是**接线缺口**:把知识任务的判据做成结构的(目标还开着 + 带着登记过的假设),系统自己进知识模式;并把图从手写 SVG 换成 React Flow,给了它一个真正的全屏工作区。
8
+
9
+ ### Added
10
+
11
+ - **知识模式(分诊)**:判据是**结构的**——目标还开着,而且它带着登记过的命题。立约(`SetGoal`)本身就是模型已经做出的承诺;普通问答从不立约,于是从不进这一档(**零成本契约**)。词面启发式猜错了没人能复核,结构判据可以。
12
+ - **知识预检**:把「已知」自动送到模型面前——只读、有界、**词面命中不猜语义**;每条读数说得出它来自哪条事实。
13
+ - **缺口读数**:从已有事实算出还缺什么形态,每条指得出一个能补的动作。真跑的反直觉结论:改变行为的其实是**缺口的可见性**,不是门——所以两者都留(可见性让它想做,门不让它绕过)。
14
+ - **知识门**:结案之前、派评估者之前拦住**没有形态的核心结论**——纯散文不许升格。
15
+ - **知识 Inspector**:点节点或边 → 定义 / 关系 / 断言 / 证据链 / 登记与修订史;「按此筛选」是详情里的**显式动作**,不猜你点它的意思。新增宿主只读路由 `/api/clearai/inspector` 与行为测试。
16
+ - **图谱工作区**:图带可展开成全屏工作区,布局是力导向(知识图谱的原生形状);渲染交给 React Flow(`@xyflow/react` 12,vendor 行随构建走,与 `dist/` 同一条纪律)。
17
+ - **哲学 P6**:「本体生长是循环的原生行为,不是另一个模式」——写进[循环哲学](docs/loop-philosophy.zh-CN.md)与[认识论循环](docs/epistemic-loop.zh-CN.md)(中英)。
18
+ - **素材工具** `tools/panel-shots.mjs`:折一场真会话 → 挂**真组件** → 真 Chrome 截图(与 `tools/graph-shots.mjs` 同一条口径)——面板截图从此可复现,不用人去界面里手点。
19
+
20
+ ### Changed
21
+
22
+ - **图 DTO 统一**:图带与工作区共用同一份投影,判据只有一处(`fold` → `view()` → `graphProjection()`);「同一份账本 ⇒ 同一张图」是投影的性质。
23
+ - **预设描述**收敛成一句中英并排:「利用认识论循环构建可信本体。Build a trustworthy ontology through the epistemic loop.」——名册只有这两行元数据,宿主不会替我们本地化,所以只能自己写死。
24
+ - **README 的「安装」改为「安装与使用」**:写明怎么在模式选择器里切到 `ClearAI`(默认标准模式不挂认识论循环),并把本体图 / 图带 / Inspector 的真机截图放进去。
25
+ - 营销物料换掉全部陈旧面板截图:知乎稿与 docx、小红书 9 张卡片改用真机会话的投影(统计同步为 15 套件 1618 条断言 · 29 个意图工具 · 6808 行内核 · 50 种事件分支)。
26
+
27
+ ### Fixed
6
28
 
7
- All notable changes to this project are recorded here. The format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
29
+ - **命题身份在修订时被重新签发**:同一句话在 id 空间里躺着两份读数(一份「已支持」、一份「未触及」),真会话的卡实测 4 条主张显示成 6~8 行。现在主张原文不变就复用原 id,换了主张才发新 id;并补上 `hypothesis/superseded` 一直缺席的生产者。
30
+ - **节点拖不动**(两次):受控 `nodes` 没接 `onNodesChange`;拖动键写成了对象,`[object Object]` 查不到。
31
+ - **React Flow 是 forwardRef 对象,不是函数**——守卫把合法组件判成「没装上」;并改为同作用域注入,不再依赖运行时模块行。
32
+ - **图带收不到 sessionId**;时间戳不再显示 1970。
33
+ - **工作区读面有主人**:子会话结构上写不进词汇 / 事实货架。
34
+ - 力导向布局的四个真机缺陷(评估者独立复核后逐条修掉)。
8
35
 
9
36
  ## [0.2.0] — 2026-09-17
10
37
 
package/README.md CHANGED
@@ -59,13 +59,26 @@ ClearAI does **not** claim recursive self-improvement. It provides the epistemic
59
59
 
60
60
  ---
61
61
 
62
- ## Install
62
+ ## Install and use
63
63
 
64
64
  ```bash
65
65
  npx clearai-dsh install
66
66
  ```
67
67
 
68
- Restart `dsh web` afterwards (`npx @deepseek-ai/dsh web`), create a session, and pick the **ClearAI** preset. If pnpm is not on PATH: `npm install -g pnpm` (do not `corepack enable` — it installs a version forwarder that may download a pnpm it cannot launch).
68
+ Restart `dsh web` afterwards (`npx @deepseek-ai/dsh web`), then **create a session and switch to the `ClearAI` mode in the picker at the top**:
69
+
70
+ 1. Open `dsh web` and click "New session";
71
+ 2. Click the current mode name at the top (default: **Standard mode**) to open the preset list;
72
+ 3. Pick **ClearAI** — its card reads "利用认识论循环构建可信本体。Build a trustworthy ontology through the epistemic loop.";
73
+ 4. Just ask your question. Ordinary Q&A runs as usual; once you set a goal and register hypotheses, the system enters knowledge mode by itself: known facts come to you, gaps stay visible, and conclusions earn their place.
74
+
75
+ <picture>
76
+ <img src="docs/shots/zh/jepa-ontology.png" alt="The knowledge graph in ClearAI mode" width="820">
77
+ </picture>
78
+
79
+ *The ontology graph in ClearAI mode — this real session grew 21 concepts and 9 predicates; the same ledger always yields the same picture. (UI shown is Chinese.)*
80
+
81
+ If pnpm is not on PATH: `npm install -g pnpm` (do not `corepack enable` — it installs a version forwarder that may download a pnpm it cannot launch).
69
82
 
70
83
  From the repository:
71
84
 
@@ -84,7 +97,19 @@ node docs/diagrams/build-hero.mjs # redraw the product hero (needs google-chro
84
97
 
85
98
  The middle column has two switchable views: **Deliverables** and **Ontology**. The right sidebar: **Worldlines** and **External Brain**.
86
99
 
87
- **Ontology** — this view is your knowledge home. At the top, a **graph band**: the ontology graph (what your domain looks like) and the entity graph (what you have actually verified) toggle with one click; clicking a concept node filters the shelves below. Below that, the **ontology shelf**: established entries, each with assertion chips (click to see what the term means), boundary, and support level; contradictions surface automatically. The vocabulary maintenance block sits collapsed at the bottom — it auto-expands when a language exists before any sentence does.
100
+ **Ontology** — this view is your knowledge home. At the top, a **graph band**: the ontology graph (what your domain looks like) and the entity graph (what you have actually verified) toggle with one click; clicking a node or edge opens the **knowledge inspector** (definition / relations / assertions / evidence chain / history), and "filter by this" is an explicit action inside the detail view. Below that, the **ontology shelf**: established entries, each with assertion chips (click to see what the term means), boundary, and support level; contradictions surface automatically. The vocabulary maintenance block sits collapsed at the bottom — it auto-expands when a language exists before any sentence does.
101
+
102
+ <picture>
103
+ <img src="docs/shots/zh/jepa-band.png" alt="The graph band: ontology graph and entity graph" width="820">
104
+ </picture>
105
+
106
+ *The band — the ontology graph and the entity graph share one deterministic projection, so the same ledger always yields the same picture (captured from a real session: 21 concepts, 9 predicates).*
107
+
108
+ <picture>
109
+ <img src="docs/shots/zh/jepa-inspector.png" alt="Knowledge inspector: definition, relations, assertions, evidence chain" width="820">
110
+ </picture>
111
+
112
+ *Open any node or edge: definition, relations, assertions, evidence chain, registration and revision history, all in one place. (UI shown is Chinese.)*
88
113
 
89
114
  **Worldlines** — when two routes genuinely disagree, they run as separate branches with their own readings; the losing one stays on record, and adoption is a human press.
90
115
 
@@ -107,6 +132,7 @@ The middle column has two switchable views: **Deliverables** and **Ontology**. T
107
132
  - [Positioning](docs/positioning.md) · [Domain ontology design](docs/domain-ontology.md)
108
133
  - [Epistemic loop](docs/epistemic-loop.md) · [Verification ontology](docs/verification-loop.md) · [Loop philosophy](docs/loop-philosophy.md)
109
134
  - [Design principles](docs/design-principles.md) · [Soul map](docs/soul-map.md) · [Glossary](docs/glossary.md)
135
+ - [Knowledge-native loop ledger](docs/optimization/knowledge-native-loop.zh-CN.md) (this round: triage / preflight / knowledge gate / graph / inspector; zh-CN)
110
136
  - [Development plan](docs/optimization/domain-ontology-plan.md) (with real-run evidence from the Hengtong project)
111
137
  - [Known gaps](docs/known-gaps.md) · [Authority map](docs/authority-map.md) · [Release verification](docs/release-verification.md)
112
138
 
package/README.zh-CN.md CHANGED
@@ -44,7 +44,7 @@ ClearAI 是一个**本体发现与探索平台**,核心由两个概念支撑
44
44
  | 完成 | 模型宣布完成 | 系统按交付的证据算出来 |
45
45
  | 裁决 | 谁做的谁说了算 | 分离——超过一定等级,做的人不能判自己 |
46
46
  | 失败 | 删掉、重来、忘掉 | 留下:被推翻的命题是结果,不是噪声 |
47
- | 沉淀 | 一段聊天记录 | **一个本体**:每条边都通过循环挣得 |
47
+ | 沉淀 | 一段聊天记录 | **一个本体**:本体中的每个关系都有来源和验证记录 |
48
48
 
49
49
  <picture>
50
50
  <source media="(prefers-color-scheme: dark)" srcset="docs/diagrams/epistemic-loop-hero-dark.zh-CN.png">
@@ -59,13 +59,26 @@ ClearAI **不**声称递归自我改进。它提供的是自我改进系统所
59
59
 
60
60
  ---
61
61
 
62
- ## 安装
62
+ ## 安装与使用
63
63
 
64
64
  ```bash
65
65
  npx clearai-dsh install
66
66
  ```
67
67
 
68
- 装完重启 `dsh web`(`npx @deepseek-ai/dsh web`),新建会话选 **ClearAI** 预设。如果 PATH 上没有 pnpm:`npm install -g pnpm`(别用 `corepack enable`——它装的是版本转发器,可能下载一个自己启动不了的 pnpm)。
68
+ 装完重启 `dsh web`(`npx @deepseek-ai/dsh web`),然后**新建会话,在顶部的模式选择器里切换到 `ClearAI`**:
69
+
70
+ 1. 打开 `dsh web`,点「新建会话」;
71
+ 2. 点顶部当前的模式名(默认是「标准模式」),展开预设列表;
72
+ 3. 选 **ClearAI**——卡片描述就一句:「利用认识论循环构建可信本体。Build a trustworthy ontology through the epistemic loop.」;
73
+ 4. 像平常一样提问。普通问答照常走;一旦你立了目标并登记假设,系统自动进入知识模式:预检把已知送上来,缺口如实读出来,结论过证据才准入。
74
+
75
+ <picture>
76
+ <img src="docs/shots/zh/jepa-ontology.png" alt="ClearAI 模式下的知识图谱" width="820">
77
+ </picture>
78
+
79
+ *ClearAI 模式下的本体图——这场真实会话长出 21 个概念、9 条谓词;同一份账本永远得到同一张图。*
80
+
81
+ 如果 PATH 上没有 pnpm:`npm install -g pnpm`(别用 `corepack enable`——它装的是版本转发器,可能下载一个自己启动不了的 pnpm)。
69
82
 
70
83
  从仓库开发:
71
84
 
@@ -84,7 +97,19 @@ node docs/diagrams/build-hero.mjs # 重画产品主图(需 google-chrome)
84
97
 
85
98
  中栏两格可切:**产物**与**本体**。右栏:**世界树**与**外脑**。
86
99
 
87
- **本体**——这一格是你的知识主场。顶部是**图带**:本体图(你的领域长什么样)与实体图(已经验证出了什么)一键切换,点概念节点按概念过滤。下面是**本体货架**:已确立的条目,每条带断言芯片(点开看这个词什么意思)、边界与等级;互相矛盾的自动亮出来。词汇维护区收在最底下——语言先于句子时它自动展开。
100
+ **本体**——这一格是你的知识主场。顶部是**图带**:本体图(你的领域长什么样)与实体图(已经验证出了什么)一键切换;点节点或边打开**知识 Inspector**(定义 / 关系 / 断言 / 证据链 / 历史),「按此过滤」是详情里的显式动作。下面是**本体货架**:已确立的条目,每条带断言芯片(点开看这个词什么意思)、边界与等级;互相矛盾的自动亮出来。词汇维护区收在最底下——语言先于句子时它自动展开。
101
+
102
+ <picture>
103
+ <img src="docs/shots/zh/jepa-band.png" alt="本体图带:本体图与实体图切换" width="820">
104
+ </picture>
105
+
106
+ *图带——本体图与实体图共用同一份确定性投影,同一份账本永远得到同一张图;素材来自一场真实会话(21 概念 · 9 谓词)。*
107
+
108
+ <picture>
109
+ <img src="docs/shots/zh/jepa-inspector.png" alt="知识 Inspector:定义、关系、断言、证据链" width="820">
110
+ </picture>
111
+
112
+ *点开任一节点或边:定义、关系、断言、证据链、登记与修订史,一处读完。*
88
113
 
89
114
  **世界树**——两条路线真的分歧时,各自独立跑、各自带读数;落选的那条留在记录里,采纳是人按的那一下。
90
115
 
@@ -107,6 +132,7 @@ node docs/diagrams/build-hero.mjs # 重画产品主图(需 google-chrome)
107
132
  - [定位](docs/positioning.zh-CN.md) · [领域本体设计](docs/domain-ontology.zh-CN.md)
108
133
  - [认识论循环](docs/epistemic-loop.zh-CN.md) · [验证本体](docs/verification-loop.zh-CN.md) · [循环哲学](docs/loop-philosophy.zh-CN.md)
109
134
  - [设计原则](docs/design-principles.zh-CN.md) · [灵魂映射](docs/soul-map.zh-CN.md) · [术语表](docs/glossary.zh-CN.md)
135
+ - [知识原生循环执行账本](docs/optimization/knowledge-native-loop.zh-CN.md)(本轮:分诊 / 预检 / 知识门 / 图 / Inspector)
110
136
  - [开发计划](docs/optimization/domain-ontology-plan.zh-CN.md)(含亨通真跑读数)
111
137
  - [已知缺口](docs/known-gaps.zh-CN.md) · [权威归属](docs/authority-map.zh-CN.md) · [发布验收](docs/release-verification.zh-CN.md)
112
138