walle45611

walle45611 / LLM-Wiki-Worker

Public

這個專案的起點很單純：我們每天都會看到很多有價值的內容，可能是一篇網頁文章、也可能是一支 YouTube 影片。當下看完覺得很有收穫，就先存進筆記裡，想著「之後一定用得到」。 但真實情況是，過了一段時間後，記憶只剩下一個模糊印象： 你知道自己「好像看過這件事」，卻說不清在哪裡看到、重點是什麼、當初為什麼重要。 LLM-Wiki Worker 就是為了解決這個而生。

19
5
100% credibility
Found Apr 23, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

This project builds an AI-powered personal knowledge base that processes notes from a shared folder into a queryable wiki, delivering answers and summaries through a chat bot.

How It Works

1
📰 Discover smart note organizer

You hear about a helpful tool that turns your daily clippings of articles and videos into a living, searchable knowledge collection.

2
📝 Gather your reads

Clip interesting web pages, articles, and video notes into organized folders in your favorite note-taking app, ready for magic.

3
🔗 Connect your world

Link your note folders, chat buddy, and clever AI thinker so they work together seamlessly behind the scenes.

4
🚀 Bring it to life

Flip the switch to launch your personal wiki helper, and it hums quietly in the background, always ready.

5
💬 Chat your questions

Send simple messages like 'What did I read today?' or 'Summarize my notes from last week' right in your chat app.

6
✨ Unlock hidden gems

Watch as smart replies arrive, pulling together summaries, highlights, and insights from your forgotten notes.

📚 Memory mastered

Your knowledge base stays fresh with auto-daily digests, letting you recall and connect ideas effortlessly forever.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 19 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is LLM-Wiki-Worker?

LLM-Wiki-Worker is a JavaScript Cloudflare Worker that turns your clipped notes from web articles or YouTube videos into an AI-maintained personal wiki on GitHub. You dump raw Markdown from Obsidian into a repo's raw/ folder, and it uses LLM agents to refine them into structured wiki/ pages with summaries, indexes, and logs—queryable via Telegram bot like "what did I read today?" or scheduled daily digests. It syncs local edits with cloud AI processing, solving the "saved it but forgot the details" problem.

Why is it gaining traction?

It stands out with rule-driven LLM agents that enforce consistent processing via GitHub file tools, plus queue-based execution to handle long AI tasks without timeouts—perfect for serverless reliability. Telegram integration delivers formatted replies with length limits and fallback safety, while cron jobs auto-generate daily wikis. Developers dig the zero-infra setup: deploy to Cloudflare, point Telegram webhook, and it runs.

Who should use this?

Obsidian power-users clipping YouTube GitHub projects, android apks, or copilot demos into notes, who want chat-based recall without manual tagging. Knowledge workers syncing GitHub repos as personal wikis via Telegram, especially those handling ios apps, desktop tools, or pages downloads. Ideal for solo devs building memory aids around LLM workflows.

Verdict

With 19 stars and 1.0% credibility score, it's early-stage but boasts solid docs, deployment scripts, and tests—deployable in minutes via wrangler. Try it if Telegram + GitHub fits your note flow; skip for production until more battle-tested.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.