@QingQ77: A collection of AI skills I use daily to auto-transcribe multi-platform content, manage knowledge bases, and monitor industry intelligence. https://github.com/chubbyguan/chubbyskills… This open-source skill pack includes 11 skills, all of which are tools the author uses daily.

Personal daily AI Skill collection: automatically transcribe content from multiple platforms, manage knowledge bases, and monitor industry intelligence. https://github.com/chubbyguan/chubbyskills… This open-source skill pack contains 11 skills, all used by the author in daily life. For video, it covers automatic transcription from Douyin, Bilibili, TikTok, Weibo, Zhihu, and YouTube. Podcasts supported include Xiaoyuzhou and Ximalaya. Also included are WeChat article processing, Obsidian knowledge base management, industry intelligence radar, and learning note automation. Each Skill is a set of instructions that an AI agent can directly load — compatible with Claude Code, Codex, OpenCode, and more. — # chubbyguan/chubbyskills Source: https://github.com/chubbyguan/chubbyskills Chinese · English # 🧰 Chubby Skills #### A collection of AI Skills I personally use every day, all open-sourced here License Skills Claude Code Codex OpenCode OpenClaw Hermes I’m Chubby, an ordinary person tinkering in the AI and content creation space. - 🐦 X/Twitter (https://x.com/Chubbyguan) - 💬 Jike (https://web.okjike.com/u/a876838d-d9a8-494b-9494-bb3410b77dd5) — All of these have been running in my own projects for a while, proven to save effort — that’s why I’m open-sourcing them. No gimmicks, just a few genuinely useful tools. Each Skill here is a structured instruction set that an Agent can directly load, following the Agent Skills (https://agentskills.io) open standard. Compatible with Claude Code, Codex, OpenCode, OpenClaw, and Hermes. — ## 📋 Table of Contents ### Video Transcription | Name | Platform | One-liner | |—|—|—| | 🎬 douyin-transcribe | Douyin | Douyin video → transcription → Markdown | | 📺 bilibili-transcribe | Bilibili | Bilibili video → transcription → Markdown | | 🎵 tiktok-transcribe | TikTok | TikTok video → transcription → Markdown | | 📱 weibo-transcribe | Weibo | Weibo video → transcription → Markdown | | 💡 zhihu-transcribe | Zhihu | Zhihu video → transcription → Markdown | | 🌍 youtube-transcribe | YouTube | YouTube → transcription → English translation → bilingual (Chinese/English) | ### Podcast Transcription | Name | Platform | One-liner | |—|—|—| | 🎙️ podcast-transcribe | Xiaoyuzhou / Ximalaya | Podcast → download → transcription → Markdown | ### Content Processing | Name | Platform | One-liner | |—|—|—| | 📰 wechat-article-ingest | WeChat Official Accounts | WeChat article link → Markdown + Layer A key points extraction + Layer B question chain | ### Knowledge Base Management | Name | One-liner | |—|—| | 🧠 knowledge-base-management | Full lifecycle management of knowledge base: three-layer architecture, material ingestion, health check, three-tool integration | ### Workflows | Name | One-liner | |—|—| | 📡 industry-intelligence-radar | Multi-source scanning (X/Jike/V2EX/HN) → keyword filtering → trend detection → daily intelligence brief | | 📚 learning-notes-automation | Video/podcast transcription → knowledge point extraction → flashcard generation → knowledge graph update | — ## 📦 Installation ### Method 1: One-click install (recommended) bash git clone https://github.com/chubbyguan/chubbyskills.git cd chubbyskills bash setup.sh # Install all dependencies bash setup.sh podcast # Install only a specific skill ### Method 2: Manual install bash git clone https://github.com/chubbyguan/chubbyskills.git cd chubbyskills pip install -r requirements.txt # All pip install -r podcast-transcribe/requirements.txt # Single skill ### Method 3: Agent install In an Agent that supports Skills (Claude Code, Codex, OpenClaw, Hermes, etc.), just say: Install this skill for me: https://github.com/chubbyguan/chubbyskills/tree/main/ > Gitee mirror: https://gitee.com/chubbyguan/chubbyskills — ## ✨ Skills ### 🎬 douyin-transcribe (Douyin Transcription) > “Want to convert Douyin videos to text? Used to spend ages dealing with cookies and yt-dlp. Now done in one command.” Douyin video → download audio → SenseVoice-Small transcription → save as Markdown. Acknowledgments: vangie/douyin-transcriber (https://github.com/vangie/douyin-transcriber) · FunAudioLLM/SenseVoice (https://github.com/FunAudioLLM/SenseVoice) → SKILL.md · Script — ### 📺 bilibili-transcribe (Bilibili Transcription) > “So many insightful videos on Bilibili — finally can convert them to text and read at my own pace.” Bilibili video → download audio via yt-dlp → SenseVoice-Small transcription → save as Markdown. Acknowledgments: yt-dlp/yt-dlp (https://github.com/yt-dlp/yt-dlp) · FunAudioLLM/SenseVoice (https://github.com/FunAudioLLM/SenseVoice) → SKILL.md · Script — ### 🎵 tiktok-transcribe (TikTok Transcription) > “Great content on TikTok can now be saved and revisited.” TikTok video → download audio → SenseVoice-Small transcription → save as Markdown. Supports vm.tiktok.com short links. Acknowledgments: yt-dlp/yt-dlp (https://github.com/yt-dlp/yt-dlp) · FunAudioLLM/SenseVoice (https://github.com/FunAudioLLM/SenseVoice) → SKILL.md · Script — ### 📱 weibo-transcribe (Weibo Video Transcription) > “Video content on Weibo — finally can convert it to text.” Weibo video → download audio → SenseVoice-Small transcription → save as Markdown. Supports weibo.com and m.weibo.cn. Acknowledgments: yt-dlp/yt-dlp (https://github.com/yt-dlp/yt-dlp) · FunAudioLLM/SenseVoice (https://github.com/FunAudioLLM/SenseVoice) → SKILL.md · Script — ### 💡 zhihu-transcribe (Zhihu Video Transcription) > “Video answers on Zhihu can now be converted to text and saved.” Zhihu video → download audio → SenseVoice-Small transcription → save as Markdown. Acknowledgments: yt-dlp/yt-dlp (https://github.com/yt-dlp/yt-dlp) · FunAudioLLM/SenseVoice (https://github.com/FunAudioLLM/SenseVoice) → SKILL.md · Script — ### 🌍 youtube-transcribe (YouTube Transcription + Translation) > “Finally understand English YouTube videos effortlessly.” YouTube video → download via yt-dlp → SenseVoice-Small transcription → automatic translation from English to Chinese → output bilingual (Chinese/English) Markdown. Acknowledgments: yt-dlp/yt-dlp (https://github.com/yt-dlp/yt-dlp) · FunAudioLLM/SenseVoice (https://github.com/FunAudioLLM/SenseVoice) · DeepSeek (https://platform.deepseek.com/) → SKILL.md · Script — ### 🎙️ podcast-transcribe (Podcast Transcription) > “Can’t finish listening to podcasts? Turn them into text and read whenever.” Podcast audio → download → faster-whisper transcription → save as Markdown. Supports Xiaoyuzhou and Ximalaya, plus RSS batch downloads. Acknowledgments: SYSTRAN/faster-whisper (https://github.com/SYSTRAN/faster-whisper) · OpenAI Whisper (https://github.com/openai/whisper) → SKILL.md · Script — ### 📰 wechat-article-ingest (WeChat Article Processing) > “WeChat official account articles become structured knowledge instantly.” WeChat official account article → Markdown extraction → Layer A key points extraction + Layer B question chain generation. Supports direct link fetching. Acknowledgments: microsoft/markitdown (https://github.com/microsoft/markitdown) · pymupdf/PyMuPDF (https://github.com/pymupdf/PyMuPDF) · dontbesilent (https://github.com/dontbesilent) → SKILL.md · Script — ### 🧠 knowledge-base-management (Knowledge Base Management) > “From material ingestion to health check — one flow covers everything.” Obsidian knowledge base full lifecycle management: three-layer architecture (Inbox/Wiki/Output), ABC-graded material ingestion, health check and cleanup, GBrain/GraphRAG/LLM Wiki three-tool integration, directory organization and archiving. Core capabilities: - 📥 Material ingestion: ABC grading + auto-sync from WeChat articles and group chats - 🔍 Health check: broken link repair, frontmatter completion, deduplication and archiving - 🛠️ Tool integration: GBrain search + GraphRAG discovery + LLM Wiki writing - 📂 Directory organization: full vault audit, batch archiving, file naming conventions Acknowledgments: Obsidian (https://obsidian.md/) · GBrain (https://github.com/) · GraphRAG (https://github.com/microsoft/graphrag) → SKILL.md · Script — ### 📡 industry-intelligence-radar (Industry Intelligence Radar) > “Know early = Act early = Gain early” Multi-source intelligence scanning system: X/Twitter + Jike + V2EX + Hacker News + 36kr → keyword filtering → trend detection → daily intelligence brief. Core capabilities: - 🔍 Multi-source parallel scanning: X, Jike, V2EX, HN, 36kr - 🏷️ Smart filtering: deduplication, timeliness, signal strength grading - 📈 Trend detection: breaking/ongoing/emerging trend identification - 📋 Intelligence brief: high-signal events + trend observations + opportunity insights Coverage areas: AI/Agent, semiconductors, aerospace, new energy, gaming, cross-border e-commerce, startups/investment → SKILL.md · Script — ### 📚 learning-notes-automation (Learning Notes Automation) > “Watching a video ≠ learning it. Generating flashcards = remembering it.” Automated processing of learning content: video/podcast transcription → knowledge point extraction → flashcard generation → knowledge graph update. Core capabilities: - 🎬 Multi-source input: YouTube, Bilibili, podcasts, Douyin, WeChat articles - 📝 Knowledge point extraction: concepts, facts, methodologies, quotes - 🃏 Flashcard generation: concept cards, Q&A cards, comparison cards, step cards (Anki compatible) - 🕸️ Knowledge graph: entity extraction + relationship mapping + auto-ingestion Output formats: - Anki flashcard file (directly importable) - Learning notes in Markdown - Knowledge base entries → SKILL.md · Script — ## 🔧 Environment Requirements ### One-click install bash bash setup.sh # All dependencies bash setup.sh podcast # Only a specific skill Or manually install requirements.txt from each skill directory. ### Video Transcription (Douyin/Bilibili/TikTok/Weibo/Zhihu/YouTube) bash pip install -r douyin-transcribe/requirements.txt # System dependencies brew install ffmpeg yt-dlp # macOS # Ubuntu: sudo apt install ffmpeg && pip install yt-dlp # YouTube translation (optional) export DEEPSEEK_API_KEY=your-key ### Podcast Transcription bash pip install -r podcast-transcribe/requirements.txt brew install ffmpeg # macOS ### WeChat Article Processing bash pip install -r wechat-article-ingest/requirements.txt ### Knowledge Base Management bash # Health check script vault_health_check.py has zero dependencies — pure Python standard library, no installation needed # The following are optional third-party tools (not bundled with this repo): # GBrain (knowledge base search) — pip install gbrain # GraphRAG (knowledge graph discovery) — independent project, see SKILL.md # LLM Wiki — integrated into the Obsidian vault ### Industry Intelligence Radar bash # scan.py has zero dependencies — pure standard library (HN + V2EX + RSS), no installation required, no API Key needed # X/Jike signals are supplemented by Agent web search (optional) ### Learning Notes Automation bash # make_notes.py requires a DeepSeek API Key for knowledge point extraction + flashcard generation export DEEPSEEK_API_KEY=your-key # Transcription step reuses the *-transcribe skills from this repo (see "Video Transcription" dependencies above) — ## 🙏 Acknowledgments ### Repository Structure Inspiration - KKKKhazix/khazix-skills (https://github.com/KKKKhazix/khazix-skills) — Repository structure and README style reference ### Speech Recognition - FunAudioLLM/SenseVoice (https://github.com/FunAudioLLM/SenseVoice) — Alibaba’s open-source Chinese speech recognition model - SYSTRAN/faster-whisper (https://github.com/SYSTRAN/faster-whisper) — CTranslate2 implementation of Whisper - OpenAI Whisper (https://github.com/openai/whisper) — Pioneering speech recognition model ### Video Download - yt-dlp/yt-dlp (https://github.com/yt-dlp/yt-dlp) — Powerful video download tool supporting thousands of platforms - vangie/douyin-transcriber (https://github.com/vangie/douyin-transcriber) — Inspiration for Douyin download approach ### Document Processing - microsoft/markitdown (https://github.com/microsoft/markitdown) — Document-to-Markdown tool - pymupdf/PyMuPDF (https://github.com/pymupdf/PyMuPDF) — PDF processing library ### Knowledge Base Management - Obsidian (https://obsidian.md/) — Knowledge base platform - GBrain (https://github.com/) — Knowledge base search tool - GraphRAG (https://github.com/microsoft/graphrag) — Knowledge graph discovery - Karpathy LLM Wiki (https://github.com/karpathy/llm-wiki) — LLM Wiki writing pattern ### AI Agent & Translation - Agent Skills (https://agentskills.io) — Open standard for Agent Skills - DeepSeek (https://platform.deepseek.com/) — LLM for translation - dontbesilent (https://github.com/dontbesilent) — A+B dual-track processing methodology - Hermes Agent (https://github.com/nousresearch/hermes-agent) — Runtime environment for these skills Thanks to all open-source contributors! 🙏 — ## 🌟 About These skills are what I personally use every day. If open-sourcing them helps you, just give a ⭐. - 🐦 X/Twitter (https://x.com/Chubbyguan) - 💬 Jike (https://web.okjike.com/u/a876838d-d9a8-494b-9494-bb3410b77dd5) — MIT License · Free to use / modify / redistribute Made by @chubbyguan (https://github.com/chubbyguan)