Your AI memory,
deployable to any AI.
One URL, every AI. Capture from ChatGPT, Claude, GitHub, Obsidian, or Notion — every doc becomes part of a hub that Claude, Cursor, ChatGPT, and Codex all fetch the same way. You set the direction; Memory.Wiki structures the URL.
Three URL scopes — Document, Bundle, Hub. Same fetch contract everywhere. See all integrations →
One URL, every surface
Write from 7 surfaces. Read from anywhere. One permanent URL connects them all.
Integrate
Your AI tools forget you between sessions. One line fixes it.
Claude Code, Cursor, Codex, and every other agent already boot with whatever sits in AGENTS.md / CLAUDE.md / .cursor/rules. Add a line that points at your Memory.Wiki bundle and the next session opens with your prior decisions, notes, and graph analysis already loaded. No API key, no vendor lock-in.
CLAUDE.md.cursor/rules/*.mdcAGENTS.mdGEMINI.md.windsurfrulesCONVENTIONS.mdWhat you can do
Every feature is built to work together. Import from anywhere, edit how you want, share with one click.
Edit directly in the rendered preview. Bold, italic, headings, lists — click and type like a word processor.
One click generates a short URL. Anyone can view your beautifully rendered document — no app needed.
Inline and display math equations rendered with LaTeX-quality precision. Write formulas that look publication-ready.
Flowcharts, sequence diagrams, Gantt charts — write them in Markdown, see them as interactive visuals.
190+ languages with syntax highlighting via highlight.js. Every programming language you write in.
Two carefully crafted themes. Your preference is saved locally. Shared docs respect the viewer's choice.
PDF, Word, PowerPoint, Excel, HTML, CSV, LaTeX, RST, RTF — drop any file and get Markdown. AI-powered Memory.Wiki restructures raw text.
Every change is tracked. Revert to any previous version. The URL stays the same — recipients always see the latest.
Customization
Combine 8 skin themes with 8 key colors to make the editor yours.
Change the entire color palette — background, text, borders, and surfaces all adapt.
Pick an accent color. Works independently on any skin theme.
Under the hood
markdown-it renders your Markdown in the browser. The editor and every viewer use the same pipeline, so what you see while writing is what readers see.
The problem
You get great answers from AI every day. Then you close the tab and they disappear forever.
Before Memory.Wiki
- - .md files are raw text nobody wants to read
- - Copy to Google Docs — formatting breaks
- - Share via Slack — looks like code
- - Different AIs, different formats
- - Edit requires VS Code or terminal
- - Version control? Manual backups
With Memory.Wiki
- + Every .md gets a beautiful, shareable URL
- + Edit in browser — no install, no login to view
- + Works with every AI — ChatGPT, Claude, Gemini
- + Edit from Mac, VS Code, Chrome, or mobile
- + Version history, sharing permissions, access control
- + Import anything — PDF, DOCX, PPTX, paste
Where Memory.Wiki is going
Everything above works today. Here is what we are building next.
Today
LivePhase 1 + 2
- Document, Bundle, Hub — three URL scopes
- AI Bundle Generation — natural language → curated collection
- Semantic search — find docs by meaning
- Multi-surface capture (Chrome, VS Code, Mac, CLI, MCP)
- WYSIWYG editing, KaTeX, Mermaid, code highlighting
- Bundle Spec v1.0 — every URL deploys to any AI
Tomorrow
Coming SoonPhase 3 — Q3 2026
- Bundle versioning — snapshots and diffs
- MCP write access — AI agents add to your hub
- LLM Wiki integration — Obsidian sync
- Format adapters — Claude XML, OpenAI Files, Gemini grounding
Beyond
VisionPhase 4+
- Bundle Spec community — open standard, multiple implementations
- Public bundle marketplace
- Team hubs and shared knowledge
- Enterprise self-host
Memory.Wiki vs Markdown publishing tools
Publish Markdown to the web. How does Memory.Wiki stack up?
| Memory.Wiki | HackMD | StackEdit | Obsidian Publish | GitHub Gist | |
|---|---|---|---|---|---|
| No signup to publish | ✓ | — | — | — | — |
| Instant shareable URL | ✓ | ✓ | — | ✓ | ✓ |
| WYSIWYG editing | ✓ | — | — | — | — |
| Source stays on your machine | ✓ | — | — | ✓ | — |
| KaTeX math | ✓ | ✓ | — | ✓ | — |
| Mermaid diagrams | ✓ | ✓ | — | ✓ | — |
| CLI publish / pipe | ✓ | — | — | — | ✓ |
| AI integration (MCP) | ✓ | — | — | — | — |
| VS Code extension | ✓ | — | — | — | — |
| Chrome extension | ✓ | — | — | — | — |
| Mac desktop app | ✓ | — | — | — | — |
| Bidirectional sync | ✓ | ✓ | — | ✓ | — |
Memory.Wiki vs AI Memory Solutions
Other tools extract memory from your conversations automatically. Memory.Wiki takes a different approach: you decide what to remember.
| Memory.Wiki | Mem0 | Letta | Notion AI | |
|---|---|---|---|---|
| Author your own memory | ✓ | — | — | △ |
| Markdown native | ✓ | — | — | △ |
| Permanent URLs | ✓ | — | — | — |
| Multi-LLM agnostic | ✓ | ✓ | ✓ | — |
| Open source engine | ✓ | ✓ | ✓ | — |
| Bundle for deployment | ✓ | — | — | — |
| Human-readable storage | ✓ | △ | △ | ✓ |
| MCP support | ✓ | ✓ | — | — |
| WYSIWYG editing | ✓ | — | — | ✓ |
| No vendor lock-in | ✓ | ✓ | ✓ | — |
Mem0 and Letta are excellent at what they do — they extract memory from your AI conversations automatically. Memory.Wiki answers a different question: what do you want to remember? You author. You bundle. You deploy.
Pricing
No Account
Use immediately, no sign-up
- + Instant rendering
- + Import / Export all formats
- + Share via hash URL
- - Local only, no cloud
- - No short URLs
Beta
Free NowEverything unlocked while we're testing
- + Unlimited documents
- + Documents never expire
- + Cloud sync across devices
- + Short URL sharing
- + AI structuring
- + All formats supported
- - Memory.Wiki badge on shared docs
Pro
After BetaPricing announced when beta ends
- + Everything in Beta
- + No badge on shared docs
- + Custom domain
- + View analytics
- + Password protection
- + Priority AI processing
- + Memory Bundles Coming Q2 2026
- + Semantic search Coming Q2 2026
- + Bundle versioning Coming Q2 2026
Build
Coming Q2 2026For AI builders and power users
- + Everything in Pro
- + Unlimited Memory Bundles
- + API access (read/write)
- + MCP server (full, write enabled)
- + Bundle versioning + snapshots
- + Bundle analytics
- + Webhook integrations
- + Format adapters (Claude XML, etc.)
The bigger picture
This is more than a markdown tool. Read why I'm building Memory.Wiki.
Read the manifesto →Start your hub.
Capture, bundle, deploy. No signup needed to try — paste any markdown and you have a Document URL in 3 seconds.
Start your hub →