knowflow 0.1.0__tar.gz

knowflow-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 KnowFlow Contributors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

knowflow-0.1.0/MANIFEST.in

@@ -0,0 +1,3 @@
+include knowflow/templates/*/plugin.yaml
+include README.md
+include LICENSE

knowflow-0.1.0/PKG-INFO

@@ -0,0 +1,264 @@
+Metadata-Version: 2.4
+Name: knowflow
+Version: 0.1.0
+Summary: Agent Knowledge Pipeline — storage-agnostic, KB-agnostic, plug-and-play pipeline that turns raw agent JSON into versioned, human-readable docs
+Author: qypvip
+Author-email: qypvip <yanpeng898@gmail.com>
+License: MIT
+Project-URL: Homepage, https://github.com/qypvip/knowflow
+Project-URL: Repository, https://github.com/qypvip/knowflow
+Project-URL: Issues, https://github.com/qypvip/knowflow/issues
+Project-URL: Changelog, https://github.com/qypvip/knowflow/commits/main
+Keywords: agent,ai-agent,knowledge-base,pipeline,llm,markdown,obsidian,automation,data-pipeline,etl
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Classifier: Topic :: Text Processing :: Markup :: Markdown
+Classifier: Typing :: Typed
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: pyyaml>=5.1
+Provides-Extra: s3
+Requires-Dist: boto3; extra == "s3"
+Provides-Extra: notion
+Requires-Dist: notion-client; extra == "notion"
+Provides-Extra: ima
+Requires-Dist: requests; extra == "ima"
+Provides-Extra: webdav
+Requires-Dist: requests; extra == "webdav"
+Provides-Extra: all
+Requires-Dist: boto3; extra == "all"
+Requires-Dist: notion-client; extra == "all"
+Requires-Dist: requests; extra == "all"
+Dynamic: author
+Dynamic: license-file
+Dynamic: requires-python
+
+# KnowFlow 🧠
+
+**Agent Knowledge Pipeline** — 存储无关、知识库无关、即装即用。
+
+让 AI Agent 产出的数据自动走完：**写入 → 版本控制 → 多端同步 → 人类可读** 的全流程。
+
+![CI](https://github.com/qypvip/knowflow/actions/workflows/ci.yml/badge.svg)
+[![PyPI](https://img.shields.io/pypi/v/knowflow.svg)](https://pypi.org/project/knowflow/)
+[![Python](https://img.shields.io/pypi/pyversions/knowflow.svg)](https://pypi.org/project/knowflow/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+
+```bash
+pip install knowflow
+
+# 从模板创建新项目
+knowflow init football-prediction --dir my-football-bot
+cd my-football-bot
+
+# Agent写入数据 → 自动转 Markdown 文档
+echo '{"match": "巴西vs阿根廷", "prediction": "巴西胜"}' > football/data/prediction.json
+knowflow run
+
+# 查看报告
+cat output/football/prediction.md
+```
+
+## 🤔 这解决什么问题
+
+如果你用 AI Agent 干活，大概率遇到过这三件事：
+
+1. **数据散落** —— Agent 每跑一次就吐几个 JSON，几十次之后你根本找不到“上周三那次分析结果在哪”。
+2. **机器看得懂，人看不懂** —— JSON 一行行挤在一起，想回顾得自己拼。
+3. **越用越乱** —— 数据只增不减，几个月后目录里几千个文件，不敢删也不知道删哪个。
+
+KnowFlow 干的就是这三件：**Agent 只管往固定目录写 JSON，剩下的（转成人类可读的 Markdown、
+自动分热/温/冷三层归档、生成索引、Git 记录每次变更）全交给它。**
+
+一句话：**Agent 负责产出，KnowFlow 负责让它留下来还有用。**
+
+## ✨ 特性
+
+| 特性 | 说明 | 状态 |
+|------|------|------|
+| 🔌 **存储无关** | Git / 本地 / S3 / 夸克网盘 / WebDAV | ✅ 已实现 |
+| 📚 **知识库无关** | 本地目录 / Obsidian | ✅ 已实现 |
+| 📦 **模板系统** | 足球预测、股票分析、日记、Wiki 等 **7 个** | ✅ 已实现 |
+| 🗄️ **自动归档** | 热→温→冷三层，数据不会无限膨胀 | ✅ 已实现 |
+| 🔄 **版本控制** | Git 后端自动记录每次变更 | ✅ 已实现 |
+| 🧪 **测试** | 68 个单元测试 + GitHub Actions | ✅ 已实现 |
+| 🔓 **开源** | MIT 协议，可商用可二次开发 | ✅ |
+
+> 仅依赖 `pyyaml`（非"零依赖"）；Notion / IMA / MkDocs 后端**尚未实现**，欢迎 PR。
+
+## 🚀 安装
+
+```bash
+pip install knowflow
+
+# 可选后端
+pip install "knowflow[s3]"       # S3 / MinIO / R2
+pip install "knowflow[all]"      # 全部后端
+```
+
+从源码安装（开发用）：
+
+```bash
+git clone https://github.com/qypvip/knowflow.git
+cd knowflow
+pip install -e .
+```
+
+国内网络拉不动 GitHub 时，可用镜像前缀：
+
+```bash
+git clone https://ghproxy.net/https://github.com/qypvip/knowflow.git
+# 或从 Gitee 克隆
+git clone git@gitee.com:qypvip/knowflow.git
+```
+
+## ⚡ 快速开始
+
+```bash
+# 查看可用模板
+knowflow list
+
+# 初始化项目
+knowflow init daily-journal --dir my-diary
+cd my-diary
+
+# 写入一些数据
+mkdir -p diary/data/entries
+cat > diary/data/entries/2026-10-02.json << 'EOF'
+{
+  "date": "2026-10-02",
+  "title": "今天学习了 KnowFlow",
+  "mood": "开心",
+  "notes": "终于有了一个统一的知识管理工具"
+}
+EOF
+
+# 运行管道（JSON → Markdown → 索引 → 归档）
+knowflow run
+
+# 查看输出
+cat output/diary/2026-10-02.md
+```
+
+## 🧪 开发与测试
+
+```bash
+pip install -e . pytest
+pytest -q            # 68 passed
+python -m knowflow list
+```
+
+CI 在 Python 3.9 / 3.11 / 3.12 上跑全量测试，另有一个"零额外依赖"冒烟任务
+验证核心模块在只装 `pyyaml` 时可 import、CLI 可用。
+
+## 📦 模板
+
+| 模板 | 命令 | 适用场景 |
+|------|------|---------|
+| ⚽ football-prediction | `knowflow init football-prediction` | 比赛预测+复盘闭环 |
+| 📈 stock-analysis | `knowflow init stock-analysis` | 多策略股票分析 |
+| 🏆 worldcup-tracker | `knowflow init worldcup-tracker` | 赛事名单追踪 |
+| 📝 daily-journal | `knowflow init daily-journal` | 个人日记+归档 |
+| 📚 project-wiki | `knowflow init project-wiki` | 项目文档自动化 |
+| 🌐 web-monitor | `knowflow init web-monitor` | 网页内容监控 |
+| 🔬 research-notes | `knowflow init research-notes` | 学术笔记管理 |
+
+## 🔌 存储后端
+
+| 后端 | 类 | 额外依赖 | 适用场景 |
+|------|----|---------|---------|
+| 🔄 `git`（默认） | `GitStorage` | git CLI | 本地 + 版本历史 |
+| 📁 `local` | `LocalStorage` | 无 | 最简，无版本 |
+| ☁️ `s3` | `S3Storage` | boto3 | AWS / MinIO / R2 |
+| 🅰️ `quark` | `QuarkStorage` | kuake_cli | 夸克网盘 |
+| 🌐 `webdav` | `WebDAVStorage` | requests | NextCloud / NAS |
+
+```bash
+knowflow init football-prediction --storage git
+knowflow init football-prediction --storage local
+```
+
+## 📚 知识库后端
+
+| 后端 | 类 |
+|------|----|
+| 📁 `local`（默认） | `LocalKB` |
+| 📝 `obsidian` | `ObsidianKB` |
+
+```bash
+knowflow init football-prediction --kb obsidian
+```
+
+Python 侧用法：
+
+```python
+from knowflow.adapters.storage import create_storage
+from knowflow.adapters.knowledge import create_knowledge   # create_kb 为别名
+
+store = create_storage("git", path="~/my-data")            # 扁平参数
+store = create_storage("git", config={"path": "~/my-data"})  # 或显式 config
+
+kb = create_knowledge("local", path="~/my-output")
+```
+
+## 🧪 命令行
+
+```bash
+knowflow list          # 列出所有模板
+knowflow init <name>   # 从模板创建项目
+knowflow run           # 运行数据管道
+knowflow status        # 查看仓库状态
+knowflow commit "msg"  # 手动提交
+knowflow archive       # 手动归档旧数据
+knowflow version       # 查看版本
+```
+
+## 🏗️ 架构
+
+```
+Agent写入数据
+    ↓
+StorageAdapter ← 可插拔（Git/本地/S3/夸克/WebDAV）
+    ↓
+Pipeline (Markdown导出→归档→索引)
+    ↓
+KnowledgeAdapter ← 可插拔（本地/Obsidian）
+```
+
+### 加一个存储后端（两步）
+
+```python
+from knowflow.core.storage import StorageAdapter
+
+class MyStorage(StorageAdapter):
+    def read(self, path): ...
+    def read_binary(self, path): ...
+    def write(self, path, content): ...
+    def write_binary(self, path, content): ...
+    def delete(self, path): ...
+    def exists(self, path): ...
+    def list(self, prefix=""): ...
+    def walk(self, prefix=""): ...
+
+from knowflow.adapters.storage import register
+register("mystorage", MyStorage)      # 完成
+```
+
+## 📄 协议
+
+MIT
+
+## 🔗 镜像
+
+- GitHub: <https://github.com/qypvip/knowflow>
+- Gitee: <https://gitee.com/qypvip/knowflow>

knowflow-0.1.0/README.md

@@ -0,0 +1,219 @@
+# KnowFlow 🧠
+
+**Agent Knowledge Pipeline** — 存储无关、知识库无关、即装即用。
+
+让 AI Agent 产出的数据自动走完：**写入 → 版本控制 → 多端同步 → 人类可读** 的全流程。
+
+![CI](https://github.com/qypvip/knowflow/actions/workflows/ci.yml/badge.svg)
+[![PyPI](https://img.shields.io/pypi/v/knowflow.svg)](https://pypi.org/project/knowflow/)
+[![Python](https://img.shields.io/pypi/pyversions/knowflow.svg)](https://pypi.org/project/knowflow/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+
+```bash
+pip install knowflow
+
+# 从模板创建新项目
+knowflow init football-prediction --dir my-football-bot
+cd my-football-bot
+
+# Agent写入数据 → 自动转 Markdown 文档
+echo '{"match": "巴西vs阿根廷", "prediction": "巴西胜"}' > football/data/prediction.json
+knowflow run
+
+# 查看报告
+cat output/football/prediction.md
+```
+
+## 🤔 这解决什么问题
+
+如果你用 AI Agent 干活，大概率遇到过这三件事：
+
+1. **数据散落** —— Agent 每跑一次就吐几个 JSON，几十次之后你根本找不到“上周三那次分析结果在哪”。
+2. **机器看得懂，人看不懂** —— JSON 一行行挤在一起，想回顾得自己拼。
+3. **越用越乱** —— 数据只增不减，几个月后目录里几千个文件，不敢删也不知道删哪个。
+
+KnowFlow 干的就是这三件：**Agent 只管往固定目录写 JSON，剩下的（转成人类可读的 Markdown、
+自动分热/温/冷三层归档、生成索引、Git 记录每次变更）全交给它。**
+
+一句话：**Agent 负责产出，KnowFlow 负责让它留下来还有用。**
+
+## ✨ 特性
+
+| 特性 | 说明 | 状态 |
+|------|------|------|
+| 🔌 **存储无关** | Git / 本地 / S3 / 夸克网盘 / WebDAV | ✅ 已实现 |
+| 📚 **知识库无关** | 本地目录 / Obsidian | ✅ 已实现 |
+| 📦 **模板系统** | 足球预测、股票分析、日记、Wiki 等 **7 个** | ✅ 已实现 |
+| 🗄️ **自动归档** | 热→温→冷三层，数据不会无限膨胀 | ✅ 已实现 |
+| 🔄 **版本控制** | Git 后端自动记录每次变更 | ✅ 已实现 |
+| 🧪 **测试** | 68 个单元测试 + GitHub Actions | ✅ 已实现 |
+| 🔓 **开源** | MIT 协议，可商用可二次开发 | ✅ |
+
+> 仅依赖 `pyyaml`（非"零依赖"）；Notion / IMA / MkDocs 后端**尚未实现**，欢迎 PR。
+
+## 🚀 安装
+
+```bash
+pip install knowflow
+
+# 可选后端
+pip install "knowflow[s3]"       # S3 / MinIO / R2
+pip install "knowflow[all]"      # 全部后端
+```
+
+从源码安装（开发用）：
+
+```bash
+git clone https://github.com/qypvip/knowflow.git
+cd knowflow
+pip install -e .
+```
+
+国内网络拉不动 GitHub 时，可用镜像前缀：
+
+```bash
+git clone https://ghproxy.net/https://github.com/qypvip/knowflow.git
+# 或从 Gitee 克隆
+git clone git@gitee.com:qypvip/knowflow.git
+```
+
+## ⚡ 快速开始
+
+```bash
+# 查看可用模板
+knowflow list
+
+# 初始化项目
+knowflow init daily-journal --dir my-diary
+cd my-diary
+
+# 写入一些数据
+mkdir -p diary/data/entries
+cat > diary/data/entries/2026-10-02.json << 'EOF'
+{
+  "date": "2026-10-02",
+  "title": "今天学习了 KnowFlow",
+  "mood": "开心",
+  "notes": "终于有了一个统一的知识管理工具"
+}
+EOF
+
+# 运行管道（JSON → Markdown → 索引 → 归档）
+knowflow run
+
+# 查看输出
+cat output/diary/2026-10-02.md
+```
+
+## 🧪 开发与测试
+
+```bash
+pip install -e . pytest
+pytest -q            # 68 passed
+python -m knowflow list
+```
+
+CI 在 Python 3.9 / 3.11 / 3.12 上跑全量测试，另有一个"零额外依赖"冒烟任务
+验证核心模块在只装 `pyyaml` 时可 import、CLI 可用。
+
+## 📦 模板
+
+| 模板 | 命令 | 适用场景 |
+|------|------|---------|
+| ⚽ football-prediction | `knowflow init football-prediction` | 比赛预测+复盘闭环 |
+| 📈 stock-analysis | `knowflow init stock-analysis` | 多策略股票分析 |
+| 🏆 worldcup-tracker | `knowflow init worldcup-tracker` | 赛事名单追踪 |
+| 📝 daily-journal | `knowflow init daily-journal` | 个人日记+归档 |
+| 📚 project-wiki | `knowflow init project-wiki` | 项目文档自动化 |
+| 🌐 web-monitor | `knowflow init web-monitor` | 网页内容监控 |
+| 🔬 research-notes | `knowflow init research-notes` | 学术笔记管理 |
+
+## 🔌 存储后端
+
+| 后端 | 类 | 额外依赖 | 适用场景 |
+|------|----|---------|---------|
+| 🔄 `git`（默认） | `GitStorage` | git CLI | 本地 + 版本历史 |
+| 📁 `local` | `LocalStorage` | 无 | 最简，无版本 |
+| ☁️ `s3` | `S3Storage` | boto3 | AWS / MinIO / R2 |
+| 🅰️ `quark` | `QuarkStorage` | kuake_cli | 夸克网盘 |
+| 🌐 `webdav` | `WebDAVStorage` | requests | NextCloud / NAS |
+
+```bash
+knowflow init football-prediction --storage git
+knowflow init football-prediction --storage local
+```
+
+## 📚 知识库后端
+
+| 后端 | 类 |
+|------|----|
+| 📁 `local`（默认） | `LocalKB` |
+| 📝 `obsidian` | `ObsidianKB` |
+
+```bash
+knowflow init football-prediction --kb obsidian
+```
+
+Python 侧用法：
+
+```python
+from knowflow.adapters.storage import create_storage
+from knowflow.adapters.knowledge import create_knowledge   # create_kb 为别名
+
+store = create_storage("git", path="~/my-data")            # 扁平参数
+store = create_storage("git", config={"path": "~/my-data"})  # 或显式 config
+
+kb = create_knowledge("local", path="~/my-output")
+```
+
+## 🧪 命令行
+
+```bash
+knowflow list          # 列出所有模板
+knowflow init <name>   # 从模板创建项目
+knowflow run           # 运行数据管道
+knowflow status        # 查看仓库状态
+knowflow commit "msg"  # 手动提交
+knowflow archive       # 手动归档旧数据
+knowflow version       # 查看版本
+```
+
+## 🏗️ 架构
+
+```
+Agent写入数据
+    ↓
+StorageAdapter ← 可插拔（Git/本地/S3/夸克/WebDAV）
+    ↓
+Pipeline (Markdown导出→归档→索引)
+    ↓
+KnowledgeAdapter ← 可插拔（本地/Obsidian）
+```
+
+### 加一个存储后端（两步）
+
+```python
+from knowflow.core.storage import StorageAdapter
+
+class MyStorage(StorageAdapter):
+    def read(self, path): ...
+    def read_binary(self, path): ...
+    def write(self, path, content): ...
+    def write_binary(self, path, content): ...
+    def delete(self, path): ...
+    def exists(self, path): ...
+    def list(self, prefix=""): ...
+    def walk(self, prefix=""): ...
+
+from knowflow.adapters.storage import register
+register("mystorage", MyStorage)      # 完成
+```
+
+## 📄 协议
+
+MIT
+
+## 🔗 镜像
+
+- GitHub: <https://github.com/qypvip/knowflow>
+- Gitee: <https://gitee.com/qypvip/knowflow>

knowflow-0.1.0/knowflow/__init__.py

@@ -0,0 +1,4 @@
+# KnowFlow
+# Agent Knowledge Pipeline — Storage-agnostic, KB-agnostic, plug-and-play
+
+__version__ = "0.1.0"

knowflow-0.1.0/knowflow/__main__.py

@@ -0,0 +1,3 @@
+"""knowflow.__main__ — Entry point for `python3 -m knowflow`"""
+from knowflow.cli import main
+main()

knowflow-0.1.0/knowflow/adapters/knowledge/__init__.py

@@ -0,0 +1,42 @@
+"""knowflow/adapters/knowledge/__init__.py — Knowledge adapter registry"""
+from typing import Optional
+
+from knowflow.core.knowledge import KnowledgeAdapter, KnowledgePage
+
+_REGISTRY = {}
+
+def register(name: str, cls):
+    _REGISTRY[name.lower()] = cls
+
+def create_knowledge(name: str, config: Optional[dict] = None, **kwargs) -> KnowledgeAdapter:
+    """Factory: create a knowledge-base adapter by name.
+
+    两种调用方式都支持：
+        create_knowledge("obsidian", path="~/my-vault")        # 扁平参数
+        create_knowledge("obsidian", config={"path": "..."})   # 显式 config 字典
+    """
+    name = name.lower()
+    if name not in _REGISTRY:
+        available = list(_REGISTRY.keys())
+        raise ValueError(
+            f"Unknown knowledge base: '{name}'. Available: {available}"
+        )
+    if config is None:
+        config = dict(kwargs) if kwargs else None
+    elif kwargs:
+        config = {**config, **kwargs}
+    return _REGISTRY[name](config)
+
+
+# 别名：core/knowledge.py 的文档里写的是 create_kb，保持一致（曾导致 ImportError）
+create_kb = create_knowledge
+
+def list_backends() -> list[str]:
+    return sorted(_REGISTRY.keys())
+
+# Built-in adapters
+from . import local as _local
+register("local", _local.LocalKB)
+
+from . import obsidian as _obsidian
+register("obsidian", _obsidian.ObsidianKB)

knowflow-0.1.0/knowflow/adapters/knowledge/local.py

@@ -0,0 +1,43 @@
+"""
+knowflow/adapters/knowledge/local.py — Local directory KB
+Simple markdown files in a directory.
+"""
+from pathlib import Path
+from knowflow.core.knowledge import KnowledgeAdapter, KnowledgePage, KnowledgeError
+
+
+class LocalKB(KnowledgeAdapter):
+    """Local markdown directory knowledge base"""
+    
+    def __init__(self, config: dict = None):
+        super().__init__(config)
+        path = config.get("path", "~/knowflow-output") if config else "~/knowflow-output"
+        self.root = Path(path).expanduser().resolve()
+        self.root.mkdir(parents=True, exist_ok=True)
+    
+    def name(self) -> str:
+        return f"Local ({self.root})"
+    
+    def sync(self, pages: list[KnowledgePage]) -> dict:
+        synced = 0
+        errors = []
+        for page in pages:
+            try:
+                target = self.root / page.path
+                target.parent.mkdir(parents=True, exist_ok=True)
+                target.write_text(page.content, encoding="utf-8")
+                synced += 1
+            except Exception as e:
+                errors.append(f"{page.path}: {e}")
+        return {"synced": synced, "skipped": 0, "errors": errors}
+    
+    def search(self, query: str, limit: int = 10) -> list[KnowledgePage]:
+        results = []
+        for md_file in sorted(self.root.rglob("*.md"))[:limit]:
+            if query.lower() in md_file.read_text(encoding="utf-8").lower():
+                results.append(KnowledgePage(
+                    title=md_file.stem,
+                    content=md_file.read_text(encoding="utf-8"),
+                    path=str(md_file.relative_to(self.root)),
+                ))
+        return results

knowflow-0.1.0/knowflow/adapters/knowledge/obsidian.py

@@ -0,0 +1,64 @@
+"""
+knowflow/adapters/knowledge/obsidian.py — Obsidian vault KB
+Syncs markdown pages to an Obsidian-compatible vault directory.
+"""
+from pathlib import Path
+from knowflow.core.knowledge import KnowledgeAdapter, KnowledgePage, KnowledgeError
+
+
+class ObsidianKB(KnowledgeAdapter):
+    """Obsidian vault knowledge base"""
+    
+    def __init__(self, config: dict = None):
+        super().__init__(config)
+        path = config.get("path", "~/obsidian-vault") if config else "~/obsidian-vault"
+        self.root = Path(path).expanduser().resolve()
+        self.root.mkdir(parents=True, exist_ok=True)
+        self._ensure_vault()
+    
+    def _ensure_vault(self):
+        """Ensure .obsidian/ config exists for vault detection"""
+        obsidian_dir = self.root / ".obsidian"
+        obsidian_dir.mkdir(exist_ok=True)
+        app_file = obsidian_dir / "app.json"
+        if not app_file.exists():
+            import json
+            app_file.write_text(json.dumps({"promptDelete": False}, indent=2))
+    
+    def name(self) -> str:
+        return f"Obsidian ({self.root})"
+    
+    def sync(self, pages: list[KnowledgePage]) -> dict:
+        synced = 0
+        errors = []
+        for page in pages:
+            try:
+                target = self.root / page.path
+                target.parent.mkdir(parents=True, exist_ok=True)
+                # Add Obsidian frontmatter tags if provided
+                content = page.content
+                if page.tags:
+                    tags_line = "\n".join(f"  - {t}" for t in page.tags)
+                    frontmatter = f"---\ntags:\n{tags_line}\n---\n\n"
+                    if not content.startswith("---"):
+                        content = frontmatter + content
+                target.write_text(content, encoding="utf-8")
+                synced += 1
+            except Exception as e:
+                errors.append(f"{page.path}: {e}")
+        return {"synced": synced, "skipped": 0, "errors": errors}
+    
+    def search(self, query: str, limit: int = 10) -> list[KnowledgePage]:
+        results = []
+        for md_file in sorted(self.root.rglob("*.md"))[:limit]:
+            try:
+                content = md_file.read_text(encoding="utf-8")
+                if query.lower() in content.lower():
+                    results.append(KnowledgePage(
+                        title=md_file.stem,
+                        content=content,
+                        path=str(md_file.relative_to(self.root)),
+                    ))
+            except Exception:
+                continue
+        return results