Scrapbook/01
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
The app and the website are in Uruguayan Spanish; the code and these docs are in English.
Contents
Related MCP server: ai-brain
Two ways to use it
Why Scrapbook/01
Share it, say what it's about
Tap Share in any app and pick Scrapbook/01. A small sheet opens on top of the app you're in, with the link filled in and the keyboard up on one question: "¿De qué se trata?" (what is it about?). Type a few words (or dictate them) and tap Save. Skipping is fine too: undescribed saves wait in an inbox.
Read more: the features tour →
It sorts itself, for a fraction of a cent
Each description goes to Jev, which picks one of your existing categories, described by what you already put in them. Only when nothing fits does Claude Haiku name a new category, in the language you wrote in. A notification says where it went, and you can move it. Sorting a save costs about $0.000014; naming a new category about $0.0002.
Read more: how saves get sorted →
A graph of everything you saved
Every save is embedded once and linked to its closest neighbors with pgvector. The graph view draws your whole library, with category hubs in Module/01's colors and similar saves linked. Each item shows its connections one or two hops out; tap one to move there while the layout stays put.
Ask your own AI about it
Scrapbook/01 is an MCP server. Add it to Claude or ChatGPT as a connector, sign in, and ask "what was that ramen place in Tokyo I saved?". It can search your library, open items and their neighbors, and save new links. Your assistant's reasoning is billed to your own plan, not to Scrapbook/01.
Host your own, for cents
Prefer your own copy? Everything here deploys to your AWS account with one command. The API and the sorting worker are Go Lambdas behind API Gateway, the web app and landing page are static sites on CloudFront, and Postgres runs on Neon, which scales to zero. With nobody using it, nothing is billed. One person's month comes to about two cents of hosting; the AI calls are paid from prepaid credits.
Read more: architecture and costs → · Host it yourself →
Run it locally in five minutes
You need Go, Bun and Docker, plus a free WorkOS sandbox for sign-in. Postgres runs in Docker, and a fake AI server stands in for Jev, the embeddings API and Claude, so the whole save, sort and link flow works without any AI keys.
Status
Early, and built in the open: the current release is v0.1.0. The Go API, the MCP server, sorting, credits and the web app are tested and run locally, and the hosted app runs on AWS behind the waitlist. Not yet proven on real hardware: the iOS share extension and push notifications on a device. Payments for credit packs aren't connected yet (packs can be granted by hand for now). Issues and pull requests are welcome.
Who's behind it
Scrapbook/01 is built by its cofounders, Franco Roura and Alexis Augoustis. It started as Alexis's waitlist page in aaugoustis/waitlist and moved here as the app grew; the full history came along.
Contributing
Bug reports, fixes and ideas are all welcome. The app is Expo (React Native) styled with the Module/01 design system; the API is Go with sqlc and River. Every pull request adds a changeset, which becomes its line in the changelog, and everyone taking part follows the code of conduct.
Read more: CONTRIBUTING.md → · Local development →
Security and license
Report security issues privately as described in SECURITY.md.
Dual-licensed under either of MIT or Apache-2.0, at your option.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Read-it-later for AI agents. Save links, bookmarks, notes, and videos as Markdown and search them.
Turn your notes, links and conversations into a Living Brain your AI tools share and build on.
Long-term memory for Claude, ChatGPT, Cursor: search, save and organize notes, links, docs, videos.
Personal wiki and memory layer for AI assistants. Persistent, structured memory across sessions.
Related MCP Servers
- AlicenseBqualityCmaintenanceEnables AI assistants like Claude to directly read, write, search, and navigate your local Logseq knowledge graph, including managing journals, pages, backlinks, and page relationships without manual copy-pasting.115 npm6MIT
- FlicenseNot gradedqualityDmaintenanceEnables AI assistants like Claude and Codex to read, write, search, and traverse Markdown notes stored in a self-hosted knowledge base.-
- AlicenseBqualityDmaintenanceConnects Memlane, your personal library of saved links and notes, to AI assistants via MCP. Enables semantic search, browsing, saving URLs, and managing notes and topics.1010 npmMIT
- FlicenseNot gradedqualityCmaintenanceEnables AI assistants to capture bookmarks, sticky notes, todos, notes, and web archives, and to search and organize personal knowledge and memory across cloud-synced clients.-