obsidian-inbox-pipeline

将任意来源(AI 资讯、经济雷达、旅游日报、RSS、文章)自动采集、 结构化写入 Obsidian inbox 的完整流水线。支持 Telegram / 飞书推送 和 cron 定时执行。一套配置,永久自动沉淀知识。

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "obsidian-inbox-pipeline" with this command: npx skills add hunanjsd/obsidian-inbox-pipeline

Obsidian Inbox Pipeline

打通外部信息源与 Obsidian 知识库的自动化流水线。安装后配置好环境变量,即可持续将资讯日报、文章、帖子等结构化沉淀到 Obsidian,无需手动整理。

核心脚本

脚本功能关键参数
capture.py写入单条笔记到 inbox--type, --title, --source, --tags, --content
query.py搜索知识库(index + 正文)--query, --type, --tags, --limit
review.py生成 inbox 整理报告--dry-run(预览)
scripts/daily_pipeline.sh三合一流水线参考radar 类型、名称、分类、来源

快速开始

1. 安装依赖

# 克隆 skill 到本地
clawhub install obsidian-inbox-pipeline

# 复制环境变量模板
cp references/.env.example .env
# 编辑 .env,填入真实路径和凭证

2. 配置环境变量

# .env 文件示例
OBSIDIAN_VAULT_PATH=/Users/you/obsidian/MyKnowledge
OBSIDIAN_INBOX_DIR=inbox
TELEGRAM_BOT_TOKEN=123456:ABC-xxx
TELEGRAM_CHAT_ID=987654321
FEISHU_APP_ID=cli_xxx
FEISHU_APP_SECRET=xxx

⚠️ 安全提醒.env 文件请加入 .gitignore,不要提交到公开仓库。脚本从不硬编码任何凭证。

3. 写入一条笔记(测试)

export OBSIDIAN_VAULT_PATH=/path/to/vault

python3 capture.py \
  --type capture \
  --title "测试笔记" \
  --source "手动测试" \
  --tags "[测试]" \
  --content "这是正文内容,可以是任何文本..."

4. 搜索知识库

export OBSIDIAN_VAULT_PATH=/path/to/vault

# 搜索标题和摘要
python3 query.py --query "OpenClaw" --limit 5

# 只看 knowledge 类型
python3 query.py --query "AI Agent" --type knowledge --limit 5

# 搜索正文内容(需要 obsidian-cli)
python3 query.py --query "OpenClaw" --include-content

5. 生成 Inbox 整理报告

export OBSIDIAN_VAULT_PATH=/path/to/vault

# 预览(不写文件)
python3 review.py --dry-run

# 正式生成(写入 notes/_review/inbox-review-YYYY-MM-DD.md)
python3 review.py

6. 定时自动运行(cron)

# 每天 7:00 自动跑(来源和 cron 触发自行配置)
0 7 * * * cd /path/to/skill && \
  OBSIDIAN_VAULT_PATH=/path/to/vault \
  TELEGRAM_BOT_TOKEN=xxx \
  TELEGRAM_CHAT_ID=xxx \
  bash scripts/daily_pipeline.sh "ai-radar" "AI 资讯雷达" "AI" "🤖" "AI 资讯" \
  >> /var/log/obsidian_pipeline.log 2>&1

流水线架构

外部来源(RSS / API / 爬虫 / 手动)
         │
         ▼
┌─────────────────────┐
│   生成原始 Markdown  │
│   (日报/文章/帖子)    │
└────────┬────────────┘
         │
         ▼
┌─────────────────────┐
│     capture.py      │◄── 写入 Obsidian inbox
│  (结构化模板写入)     │    (type / tags / source / TL;DR)
└────────┬────────────┘
         │
         ▼
┌─────────────────────┐
│  rebuild_index.mjs  │◄── 重建 .ai/index.json
│  (Obsidian 官方脚本)  │    (支持 query.py 搜索)
└────────┬────────────┘
         │
         ▼
┌─────────────────────┐
│  query.py / review  │◄── 知识库检索 / inbox 整理
│                     │
└─────────────────────┘

可选通知渠道(均走环境变量,不硬编码):
  ├── Telegram Bot(需 TELEGRAM_BOT_TOKEN)
  └── 飞书自建应用(需 FEISHU_APP_ID / FEISHU_APP_SECRET)

frontmatter 标准格式

写入的每条笔记使用统一 frontmatter,便于后续检索和分类:

---
type: capture           # capture | daily-report | knowledge
created: 2026-03-21     # 自动时间戳
tags: [AI, 日报]        # 自定义标签
source: AI 资讯雷达     # 来源
status: inbox           # inbox → knowledge → _archive
---

配合 AI 雷达使用

配合 ai-radareconomy-radartravel-radar 等 skill 使用时:

# 三合一每日自动运行(配合 cron)
0 7 * * * \
  source /path/to/.env && \
  bash /path/to/obsidian-inbox-pipeline/scripts/daily_pipeline.sh \
    "ai-radar" "AI 资讯雷达" "AI" "🤖" "AI 资讯" && \
  bash /path/to/obsidian-inbox-pipeline/scripts/daily_pipeline.sh \
    "economy-radar" "经济雷达" "经济" "📊" "经济资讯" && \
  bash /path/to/obsidian-inbox-pipeline/scripts/daily_pipeline.sh \
    "travel-radar" "旅游雷达" "旅游" "✈️" "旅游资讯"

故障排查

问题解决方案
OBSIDIAN_VAULT_PATH 报错确保环境变量已设置,或安装 obsidian-cli
inbox 写入了但 query 搜不到运行 node /path/to/vault/scripts/rebuild_index.mjs 重建索引
Telegram 推送失败确认 Bot Token 和 Chat ID 正确,Bot 已被用户发起过对话
飞书推送失败检查 app_id / app_secret 权限(需要发消息权限)

依赖

  • Python ≥ 3.8
  • obsidian-cli(可选,vault 路径自动发现)
  • node(可选,索引重建脚本)
  • Telegram Bot Token(可选,通知推送)
  • 飞书自建应用凭证(可选,通知推送)

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

通义晓蜜 - 智能外呼

触发阿里云晓蜜外呼机器人任务,自动批量拨打电话。适用于批量外呼、客户回访、满意度调查、简历筛查约面试等场景。可从前置工具或节点获取外呼名单。

Registry SourceRecently Updated
General

Letterboxd Watchlist

Scrape a public Letterboxd user's watchlist into a CSV/JSONL list of titles and film URLs without logging in. Use when a user asks to export, scrape, or mirror a Letterboxd watchlist, or to build watch-next queues.

Registry SourceRecently Updated
General

Seedance Video Generation

Generate AI videos using ByteDance Seedance. Use when the user wants to: (1) generate videos from text prompts, (2) generate videos from images (first frame, first+last frame, reference images), or (3) query/manage video generation tasks. Supports Seedance 1.5 Pro (with audio), 1.0 Pro, 1.0 Pro Fast, and 1.0 Lite models.

Registry SourceRecently Updated
4.2K17jackycser
General

Universal Skills Manager

The master coordinator for AI skills. Discovers skills from multiple sources (SkillsMP.com, SkillHub, and ClawHub), manages installation, and synchronization...

Registry SourceRecently Updated