AyanbekDos

AyanbekDos / memoriki

Public

Memoriki - LLM Wiki + MemPalace. Personal knowledge base with real memory.

claude-code karpathy knowledge-base llm mcp
34
4
100% credibility
Found Apr 10, 2026 at 34 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

Memoriki is a personal knowledge base system that uses AI to automatically structure notes and sources into a wiki with semantic search and relationship graphs.

How It Works

1
📚 Discover Memoriki

You hear about a smart tool that turns your scattered notes and articles into an organized personal knowledge library.

2
🏠 Set up your space

You create a cozy folder on your computer to hold all your knowledge.

3
🔍 Add smart search

You enable a special search feature that understands the meaning behind your content.

4
🤖 Connect your AI helper

You link an AI assistant that will read, summarize, and connect everything for you.

5
📄 Add your content

You drop articles, notes, or transcripts into your sources area.

6
✨ Let AI organize it

You ask the AI to read your new content and weave it into a structured wiki with links and summaries.

7
❓ Ask questions

You search or query naturally, and the AI pulls together answers from your entire collection.

🎉 Your living knowledge base

You now have a growing, intelligent second brain that remembers, connects ideas, and helps you discover insights effortlessly.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 34 to 34 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 memoriki?

Memoriki builds a personal knowledge base that fuses an LLM-powered wiki for structured notes with MemPalace for semantic search and entity graphs, creating real memory over scattered files. Drop articles or transcripts into a raw folder, ingest them via an MCP-compatible LLM agent like Claude Code, and get auto-generated wiki pages, cross-references, and synthesis. It's Python-based, leveraging Markdown wiki-links and MemPalace's MCP server to turn raw content into a searchable, evolving base—no more grep-style hunting.

Why is it gaining traction?

It stands out by layering wiki structure on semantic search, avoiding the "library without a catalog" pitfall of plain LLM wikis or empty searches in raw chunk tools. Developers hook on the quick start: clone, pip install mempalace, init, connect MCP, and chat-ingest files for instant query power plus linting for gaps. The compounding synthesis—entities, concepts, timestamps—feels like a memory kinder that grows smarter, not a memory killer dumping data.

Who should use this?

Founders tracking customer interviews, competitors, and pivots in one wiki. Researchers ingesting papers for cross-analysis. Students organizing lecture notes into a second brain with semantic recall. Small teams centralizing meeting threads and decisions via AI-maintained pages.

Verdict

Try it for a proof-of-concept personal wiki if you're deep into Claude or MCP agents—setup is dead simple, docs are solid for 34 stars. But with 1.0% credibility score and early maturity, expect tweaks; it's raw potential, not battle-tested production.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.