Setup & Installation
openclaw skills install @ryanhong666/knowledge-base-collector
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
npx clawhub install knowledge-base-collector
What This Skill Does
Ingests web URLs, X/Twitter links, WeChat Official Account articles, and screenshots into a structured personal knowledge base. Each item is stored as a markdown file with metadata and tags, indexed for search. Includes a fallback to a connected macOS node for WeChat articles blocked by cloud fetch restrictions.
Handles WeChat's bot-detection blocks via a local macOS node fallback, which generic bookmarking tools cannot do.
When to use it
- Saving a WeChat article for later reference
- Archiving a Twitter thread with custom tags
- Ingesting a screenshot of documentation via OCR
- Searching saved articles by tag through Telegram
- Generating a weekly digest from recently saved links
Example Workflow
Here's how your AI assistant might use this skill in practice.
INPUT
User sends: https://mp.weixin.qq.com/s/example with note 'about RAG pipelines'
AGENT
- 1Identifies URL as a WeChat Official Account link
- 2Attempts cloud fetch via r.jina.ai, which returns a verification block
- 3Routes request to connected macOS node (Reed-Mac) for local fetch
- 4Writes content.md and meta.json under kb/20_Inbox/urls/YYYY-MM/<item>/
- 5Appends entry to index.jsonl with tags #rag #wechat
OUTPUT
Article stored and indexed, searchable via Telegram with query 'rag pipelines'