Skip to content

Hypergraph is more powerful. Transform unstructured text into structured knowledge with LLMs. Graphs, hypergraphs, and spatio-temporal extractions — with one command.

docs: fold What's News, highlight provenance in features, update home cards

yifanfeng976d8811cc8f64
diff --git a/README.md b/README.md
index cc6b202..1021f13 100644
--- a/README.md
+++ b/README.md
@@ -64,41 +64,27 @@
- **📁 Per-File Source Attribution** — `he parse ./docs/` attributes each file by its name automatically; roll back or audit any single file later. An explicit `--source` still overrides.
- **⏱️ Spatiotemporal Provenance** — temporal/spatial/spatio-temporal graphs fully support source attribution and rollback, with deterministic (MERGE_FIELD) replay tests.
-### v0.7.0
-
-- **🗑️ Two-Tier Knowledge Deletion** — `he remove --node/--edge` hard-deletes by key; `he remove --edit-node --fact` removes a single wrong fact via LLM rewrite (key-invariance check, dry-run, automatic backups). *(#84)*
-- **📜 Source Attribution & Provenance** — `he feed --source` / `he parse --source` record each document's raw contributions; `he remove --document` rolls back exactly what one document contributed, re-merging keys shared with surviving sources. *(#84)*
-- **📈 Incremental Everything** — feed/parse/removal/edit now patch the vector index in place (only affected vectors re-embedded); `he feed` skips documents whose content hash is unchanged (`--refeed` to force). *(#84)*
-- **👁️ `he info --sources`** — see which documents contributed to a knowledge abstract and how much. *(#84)*
-- **🧪 `he template validate`** — catch semantic template errors before paying for LLM calls: 9 diagnostic rules, `--json` for CI. *(#77)*
-- **📊 GraphML & CSV Export** + **🌐 OrcaRouter provider** + **🔐 config.toml saved 0600** + **🔗 Obsidian wikilink fix**. *(#85, #71, #86, #87)*
-
-### v0.6.0
-
-- **🗑️ Knowledge Deletion (two-tier)** — `he remove` / `ka.remove_nodes()` hard-delete by key (orphan edges pruned); `ka.edit_node()` removes a single wrong fact via LLM rewrite, with dry-run, key-invariance checks, and automatic backups. Graph, hypergraph, and temporal/spatial KAs. *(#84)*
-- **🧪 `he template validate`** — catch semantic template errors (missing `time_field`, unknown identifier fields, bad placeholders…) before calling an LLM: 9 diagnostic rules, `--json` for CI, `--all` for directories. *(#77)*
-- **📊 GraphML & CSV Export** — `he export graphml` for desktop graph tools, `he export csv` for spreadsheets (hypergraphs get a hyperedges table). *(#85)*
-- **🔐 Config File Permissions** — `~/.he/config.toml` is now saved `0600` (owner-only) instead of world-readable. *(#86)*
-- **🔗 Obsidian Link Fix** — wikilink aliases no longer break on titles containing `[ ] | # ^`. *(#87)*
-
-### v0.5.0
-
-- **🛡️ Chunk-Level Fault Isolation** — One failed chunk (rate limit, timeout, unparseable output) no longer discards the rest of a multi-chunk extraction. Graph-family extraction degrades to a partial result + warning. *(#78)*
-- **⚡ MCP Python SDK 2.x Support** — `he-mcp` now works with both mcp 1.x and 2.x; `hyperextract[mcp]` resolves `mcp<3` (protocol 2026-07-28). *(#72, #82)*
-- **🔀 Directed-Edge Fix** — `relation_members` now preserves `(source, target)` direction instead of sorting endpoints, so A→B and B→A stay distinct edges. Custom endpoint field names are supported. *(#74)*
-- **🔑 DeepSeek API Key Fix** — `create_llm("deepseek")` now reads `DEEPSEEK_API_KEY` on the OpenAI-compatible path. *(#76)*
-- **🎓 Education Templates** — Two new presets: `education/course_concept_graph` and `education/curriculum_structure`. *(#80)*
-- **🌐 OrcaRouter Provider** — One key for 150+ models (OpenAI, Anthropic, Google, DeepSeek, Qwen…) via `create_client("orcarouter")`. *(#71)*
-- **🧭 Smaller Fixes** — `Graph_RAG.search` returns a consistent 3-tuple in default mode; `he talk -i` honors `--top-k`; clearer docs on onboarding, dev install, and structured output. *(#70, #73, #57)*
-
-### Earlier
-
-- **🤖 DeepSeek Provider** — Use `deepseek-v4-flash` / `deepseek-v4-pro` directly via `create_client(llm="deepseek")`. V4 thinking mode is auto-disabled for structured extraction compatibility. *(#67, v0.4.0)*
-- **🔌 MCP Server** — Query your knowledge abstracts from Claude Desktop and IDE agents with `he-mcp`. *(PR #40)*
-- **🧠 Anthropic Claude Support** — Use `claude-opus-4-8`, `claude-sonnet-4-6`, and `claude-haiku-4-5` directly as your LLM provider. *(PR #38)*
-- **📝 Obsidian Export** — Turn any graph into an Obsidian vault with Markdown notes linked by `[[wikilinks]]`. *(PR #37)*
-- **🧹 `he clean`** — Remove a KA's index or the whole knowledge abstract in one command. *(PR #39)*
-- **🔧 Reliability Fixes** — True mean for multi-chunk embeddings, capped OpenAI-compatible batch sizes, and resolved multi-word `llm_*` merge strategies. *(PRs #35, #36, #41)*
+<details>
+<summary><b>v0.5.0 – v0.7.0</b> — provenance, deletion, incremental index, template validator, GraphML/CSV export</summary>
+
+- **🗑️ Two-Tier Knowledge Deletion** — hard-delete by key (`he remove --node/--edge`, orphan edges pruned) or remove a single wrong fact via LLM-assisted editing (`he remove --edit-node --fact`), with dry-run, key-invariance checks, and automatic backups. *(#84)*
+- **📜 Source Attribution & Provenance** — `he feed --source` / `he parse --source` record each document's raw contributions; `he remove --document` rolls back exactly what one document contributed; `he info --sources` shows the ledger. *(#84)*
+- **📈 Incremental Everything** — feed/parse/removal/edit patch the vector index in place (only affected vectors re-embedded); `he feed` skips documents whose content hash is unchanged (`--refeed` to force). *(#84)*
+- **🧪 `he template validate`** — catch semantic template errors before paying for LLM calls: 9 diagnostic rules, `--json` for CI, `--all` for directories. *(#77)*
+- **📊 GraphML & CSV Export** — desktop graph tools and spreadsheets; hypergraphs get a hyperedges table. *(#85)*
+- **🌐 OrcaRouter Provider** — one key for 150+ models via `create_client("orcarouter")`. *(#71)*
+- **🔐 Config File Permissions** — `~/.he/config.toml` saved `0600`. *(#86)*
+- **🔗 Obsidian wikilink fix** — aliases no longer break on `[ ] | # ^`. *(#87)*
+- **🛡️ Chunk-Level Fault Isolation** — one failed chunk no longer discards the rest of a multi-chunk extraction. *(#78)*
+- **⚡ MCP Python SDK 2.x** — `he-mcp` works on mcp 1.x and 2.x. *(#72, #82)*
+- **🔀 Directed-Edge Fix** — `(source, target)` order preserved; custom endpoint field names. *(#74)*
+- **🔑 DeepSeek API Key Fix** — `DEEPSEEK_API_KEY` honored on the OpenAI-compatible path. *(#76)*
+- **🎓 Education Templates** — `course_concept_graph` + `curriculum_structure`. *(#80)*
+- **🧭 Smaller Fixes** — Graph_RAG.search 3-tuple; `he talk -i --top-k`; onboarding/docs overhaul. *(#70, #73, #57)*
+
+</details>
+
+### Archived
See the full changelog in the [GitHub releases](https://github.com/yifanfeng97/hyper-extract/releases).
@@ -111,7 +97,7 @@ Hyper-Extract is an intelligent, LLM-powered knowledge extraction and evolution
| 🔷 **8 Knowledge Structures** | From simple Lists to advanced Graphs, Hypergraphs, and Spatio-Temporal Graphs |
| 🧠 **10+ Extraction Engines** | GraphRAG, LightRAG, Hyper-RAG, KG-Gen, and more — ready to use |
| 📝 **80+ YAML Templates** | Zero-code extraction across Finance, Legal, Medical, TCM, Industry, and General domains |
-| 🔄 **Incremental Evolution** | Feed new documents anytime, and remove or refine knowledge (`he remove`) as your sources change |
+| 🔄 **Incremental Evolution & Provenance** | Feed new documents anytime — every source is attributed and the index updates incrementally; audit (`he info --sources`), roll back (`he remove --document`), or upsert updated versions as your sources change |
| 📤 **Obsidian Export** | Turn any extracted graph into an Obsidian vault — Markdown notes linked by `[[wikilinks]]` |
## 🎯 What Can You Do With It?
diff --git a/README_ZH.md b/README_ZH.md
index 32889fb..854fd09 100644
--- a/README_ZH.md
+++ b/README_ZH.md
@@ -64,41 +64,27 @@
- **📁 逐文件来源标注** — `he parse ./docs/` 自动按文件名标注每个文件;之后可单独回滚或审计任意文件。显式 `--source` 仍可覆盖。*(#84)*
- **⏱️ 时空图谱溯源** — 时序/空间/时空图谱完整支持来源标注与回滚,MERGE_FIELD 重放测试保证确定性。*(#84)*
-### v0.7.0
+<details>
+<summary><b>v0.5.0 – v0.7.0</b> — 溯源、删除、增量索引、模板校验、GraphML/CSV 导出</summary>
-- **🗑️ 两级知识删除** — `he remove --node/--edge` 按 key 硬删除;`he remove --edit-node --fact` 借助大模型移除单条错误事实(key 不变性校验、dry-run、自动备份)。*(#84)*
-- **📜 来源标注与溯源** — `he feed --source` / `he parse --source` 记录每份文档的原始贡献;`he remove --document` 精确回滚单份文档的贡献,共享 key 从幸存来源重新合并。*(#84)*
+- **🗑️ 两级知识删除** — 按 key 硬删除(`he remove --node/--edge`,孤儿边自动清理),或借助大模型移除单条错误事实(`he remove --edit-node --fact`),支持 dry-run、key 不变性校验与自动备份。*(#84)*
+- **📜 来源标注与溯源** — `he feed --source` / `he parse --source` 记录每份文档的原始贡献;`he remove --document` 精确回滚单份文档的贡献;`he info --sources` 查看来源账本。*(#84)*
- **📈 全链路增量** — feed/parse/删除/编辑均原地修补向量索引(只重嵌受影响向量);`he feed` 对内容哈希未变的文档自动跳过(`--refeed` 强制重喂)。*(#84)*
-- **👁️ `he info --sources`** — 查看知识摘要的文档贡献者及贡献量。*(#84)*
-- **🧪 `he template validate`** — 在调用大模型前发现模板语义错误:9 条诊断规则、`--json` 接入 CI。*(#77)*
-- **📊 GraphML 与 CSV 导出** + **🌐 OrcaRouter 提供商** + **🔐 config.toml 0600 权限** + **🔗 Obsidian 链接修复**。*(#85, #71, #86, #87)*
-
-### v0.6.0
-
-- **🗑️ 知识删除(两级)** — `he remove` / `ka.remove_nodes()` 按 key 硬删除(孤儿边自动清理);`ka.edit_node()` 借助大模型移除单条错误事实,支持 dry-run 预览、key 不变性校验和自动备份。覆盖 graph、hypergraph 及时序/空间图谱。*(#84)*
-- **🧪 `he template validate`** — 在调用大模型之前发现模板语义错误(缺 `time_field`、引用不存在的字段、占位符错误等):9 条诊断规则、`--json` 便于接入 CI、`--all` 批量校验目录。*(#77)*
-- **📊 GraphML 与 CSV 导出** — `he export graphml` 导入桌面图分析工具,`he export csv` 导入表格软件(超图会额外生成超边表)。*(#85)*
-- **🔐 配置文件权限** — `~/.he/config.toml` 现在以 `0600`(仅所有者可读写)保存,而非全局可读。*(#86)*
-- **🔗 Obsidian 链接修复** — wikilink 别名不再因标题包含 `[ ] | # ^` 而断链。*(#87)*
+- **🧪 `he template validate`** — 在调用大模型前发现模板语义错误:9 条诊断规则、`--json` 接入 CI、`--all` 批量校验。*(#77)*
+- **📊 GraphML 与 CSV 导出** — 对接桌面图分析工具与表格软件;超图附带超边表。*(#85)*
+- **🌐 OrcaRouter 提供商** — 一个 key 调用 150+ 模型:`create_client("orcarouter")`。*(#71)*
+- **🔐 配置文件权限** — `~/.he/config.toml` 以 `0600` 保存。*(#86)*
+- **🔗 Obsidian 链接修复** — wikilink 别名不再因 `[ ] | # ^` 断链。*(#87)*
+- **🛡️ Chunk 级故障隔离** — 单个 chunk 失败不再丢弃多 chunk 抽取的其余结果。*(#78)*
+- **⚡ MCP Python SDK 2.x** — `he-mcp` 同时兼容 mcp 1.x / 2.x。*(#72, #82)*
+- **🔀 有向边修复** — 保留 `(source, target)` 方向;支持自定义端点字段名。*(#74)*
+- **🔑 DeepSeek API Key 修复** — OpenAI 兼容路径正确读取 `DEEPSEEK_API_KEY`。*(#76)*
+- **🎓 教育领域模板** — `course_concept_graph` 与 `curriculum_structure`。*(#80)*
+- **🧭 其他修复** — Graph_RAG.search 3-tuple;`he talk -i` 支持 `--top-k`;入门/开发安装/结构化输出文档更新。*(#70, #73, #57)*
-### v0.5.0
-
-- **🛡️ Chunk 级故障隔离** — 多 chunk 抽取中,单个 chunk 失败(限流、超时、解析错误)不再导致整个任务失败并丢弃其余数据,而是降级为部分结果 + 警告。*(#78)*
-- **⚡ 支持 MCP Python SDK 2.x** — `he-mcp` 同时兼容 mcp 1.x / 2.x;`hyperextract[mcp]` 放宽到 `mcp<3`(协议 2026-07-28)。*(#72, #82)*
-- **🔀 有向边修复** — `relation_members` 保留 `(source, target)` 方向,不再排序端点,A→B 与 B→A 不再被误合并;支持自定义端点字段名。*(#74)*
-- **🔑 DeepSeek API Key 修复** — `create_llm("deepseek")` 在 OpenAI 兼容路径上正确读取 `DEEPSEEK_API_KEY`。*(#76)*
-- **🎓 教育领域模板** — 新增 `education/course_concept_graph` 与 `education/curriculum_structure` 两个预设。*(#80)*
-- **🌐 OrcaRouter 提供商** — 一个 key 调用 150+ 模型(OpenAI、Anthropic、Google、DeepSeek、Qwen…):`create_client("orcarouter")`。*(#71)*
-- **🧭 其他修复** — `Graph_RAG.search` 默认模式返回一致的 3-tuple;`he talk -i` 支持 `--top-k`;入门、开发安装与结构化输出文档更新。*(#70, #73, #57)*
-
-### 更早
+</details>
-- **🤖 DeepSeek 提供商** — 直接通过 `create_client(llm="deepseek")` 使用 `deepseek-v4-flash` / `deepseek-v4-pro`。V4 thinking 模式自动关闭,保证结构化抽取兼容。*(#67, v0.4.0)*
-- **🔌 MCP 服务器** — 通过 `he-mcp` 在 Claude Desktop 和 IDE 智能体中查询你的知识摘要。*(PR #40)*
-- **🧠 Anthropic Claude 支持** — 直接调用 `claude-opus-4-8`、`claude-sonnet-4-6`、`claude-haiku-4-5` 作为 LLM 提供商。*(PR #38)*
-- **📝 Obsidian 导出** — 将任意图谱导出为 Obsidian 知识库,Markdown 笔记通过 `[[双向链接]]` 关联。*(PR #37)*
-- **🧹 `he clean` 命令** — 一条命令清理知识摘要的索引或整个 KA。*(PR #39)*
-- **🔧 稳定性修复** — 多 chunk 嵌入采用真实均值、限制 OpenAI 兼容接口的批处理大小、修复多词 `llm_*` 合并策略。*(PRs #35、#36、#41)*
+### 更早版本
完整更新日志请参阅 [GitHub releases](https://github.com/yifanfeng97/hyper-extract/releases)。
@@ -111,7 +97,7 @@ Hyper-Extract 是一个智能的、由大语言模型(LLM)驱动的知识提
| 🔷 **8 种知识结构** | 从简单的列表到复杂的图谱、超图、时空图谱 |
| 🧠 **10+ 提取引擎** | GraphRAG、LightRAG、Hyper-RAG、KG-Gen 等开箱即用 |
| 📝 **80+ YAML 模板** | 零代码提取,覆盖金融、法律、医疗、中医、工业、通用 6 大领域 |
-| 🔄 **增量演进** | 随时喂入新文档扩展知识库,也支持删除或修正已有知识(`he remove`) |
+| 🔄 **增量演进与溯源** | 随时喂入新文档——每个来源自动标注、索引增量更新;可审计(`he info --sources`)、回滚(`he remove --document`)或以 upsert 方式更新文档 |
| 📤 **Obsidian 导出** | 将提取的图谱导出为 Obsidian 知识库——以 `[[双向链接]]` 关联的 Markdown 笔记 |
## 🎯 它能做什么?
diff --git a/docs/en/index.md b/docs/en/index.md
index 4926e10..e8c11a0 100644
--- a/docs/en/index.md
+++ b/docs/en/index.md
@@ -68,11 +68,11 @@
Ready-to-use templates for Finance, Legal, Medical, TCM, and Industry. Zero configuration needed.
-- :material-sync:{ .lg .middle } **Incremental Evolution**
+- :material-sync:{ .lg .middle } **Incremental Evolution & Provenance**
---
- Feed new documents to continuously expand your knowledge abstract. No need to reprocess everything.
+ Feed new documents to continuously expand your knowledge abstract — every source is attributed, and removal or rollback is incremental too. No full reprocessing, ever.
</div>
diff --git a/docs/zh/index.md b/docs/zh/index.md
index 6081b1e..5b80554 100644
--- a/docs/zh/index.md
+++ b/docs/zh/index.md
@@ -68,64 +68,18 @@
开箱即用的金融、法律、医疗、中医和工业模板。无需配置即可使用。
-- :material-sync:{ .lg .middle } **增量演进**
+- :material-sync:{ .lg .middle } **增量演化与溯源**
---
- 向知识库持续添加新文档。无需重新处理所有内容。
+ 持续喂入新文档扩展知识摘要——每个来源自动标注,删除与回滚同样增量完成,无需全量重处理。
-</div>
-
----
-
-## 🎯 选择您的路径
-
-<div class="grid cards" markdown>
-
-- :material-console:{ .lg .middle } __CLI 用户__
-
- ---
-
- 直接从终端处理文档。非常适合:
-
- - 快速知识提取
- - 批量文档处理
- - 无需编码构建知识库
-
- [:octicons-arrow-right-24: CLI 指南](cli/index.md)
-
-- :material-language-python:{ .lg .middle } __Python 开发者__
-
- ---
-
- 集成到您的 Python 应用程序中。非常适合:
-
- - 自定义提取管道
- - 与现有工作流程集成
- - 构建 AI 驱动的应用程序
-
- [:octicons-arrow-right-24: Python SDK](python/index.md)
-
-- :material-school:{ .lg .middle } __想了解更多?__
+- :material-category:{ .lg .middle } **8 种知识结构**
---
- 了解核心概念和架构:
-
- - 自动类型如何工作
- - 选择提取方法
- - 创建自定义模板
-
- [:octicons-arrow-right-24: 核心概念](concepts/index.md)
-
-</div>
-
----
-
-## 🧩 8 种自动类型一览
+ 从简单集合到图谱、超图与时空图谱,按需选择最合适的数据结构。
-| 类型 | 用例 | 示例输出 |
-|------|----------|----------------|
| **AutoModel** | 结构化摘要 | 带有特定字段的 Pydantic 模型 |
| **AutoList** | 项目集合 | 实体或事实的列表 |
| **AutoSet** | 去重集合 | 唯一项目的集合 |
diff --git a/uv.lock b/uv.lock
index 0fef7b3..14ad6e2 100644
--- a/uv.lock
+++ b/uv.lock
@@ -903,7 +903,7 @@ wheels = [
[[package]]
name = "hyperextract"
-version = "0.7.0"
+version = "0.8.0"
source = { editable = "." }
dependencies = [
{ name = "faiss-cpu" },