Frequently asked questions

Unabyss is a personal context layer for AI tools. It connects to your sources -- LinkedIn, Gmail, Notion, GitHub, and others -- and extracts a structured context vault that captures who you are, what you're working on, and how you communicate. That context is then served to any AI tool you use -- automatically via MCP, or via exports you can paste anywhere -- so Claude, Cursor, and other agents know who you are before the first message, without you explaining yourself every time.

Every new account starts with $25 in free credits from us -- enough to explore the product and get up and running. After that, Unabyss is pay-as-you-go: you pay for what you use, with no monthly subscription required. No credit card needed to start.

Under 3 minutes. Connect the sources you want to start with -- Unabyss will extract your context stored there. Then connect Unabyss to Claude Desktop or Cursor via MCP, or export your context files directly and paste them into any AI tool.

You connect the sources you want to start with. Available at launch: LinkedIn, X, your personal website, Gmail, Notion, Slack, Google Docs, Google Calendar, Google Drive, GitHub, and Obsidian.

You can also import existing context from ChatGPT custom instructions and Claude Projects -- so context you've already built elsewhere isn't lost.

A personal context layer is a structured, portable store of your identity, professional background, and preferences that any AI tool can access -- with your permission -- before the first interaction begins. Unlike AI memory (which is built reactively from past conversations and locked to one platform), a context layer is intentional and pre-extracted from authoritative sources like LinkedIn, Notion, or Gmail. Unabyss is the context layer: the place where your context lives, independent of any single AI tool.

AI memory is what a platform learns about you from past conversations. It's reactive, unstructured, and locked to the platform that built it -- switch tools and you start from zero. AI context is intentional and pre-extracted: who you are, structured from authoritative sources, ready to serve to any tool before the first message. Memory is a side effect of using AI. Context is infrastructure you own.

There are two fundamental differences: source and portability.

ChatGPT Memory and Claude Memory are reactive -- they learn about you from past conversations, one interaction at a time. They start empty and build slowly. Unabyss is pre-extracted -- it pulls structured context from your actual sources before any interaction begins. You start with something accurate, not a blank page.

The second difference is portability. ChatGPT Memory lives inside ChatGPT. Claude Memory lives inside Claude. If you switch tools, your context doesn't travel with you. Unabyss is cross-platform by design -- the same context vault serves Claude, Cursor, ChatGPT, and any other MCP-compatible tool. You own it. It goes where you go.

Many power users already maintain .md files with their bio, company info, and preferences. The problem is maintenance -- every time your role changes, your projects shift, or your goals evolve, you have to manually update every file across every tool.

Unabyss keeps your context vault automatically up to date. Because it stays connected to your sources, changes propagate without any manual work. And because it pulls from multiple sources simultaneously -- LinkedIn, Gmail, Notion, GitHub -- it cross-references facts across all of them, catching inconsistencies and building richer, more accurate context than any single source alone could provide. You get the portability of .md files with none of the maintenance.

You don't have to. Unabyss stays connected to your sources and updates your context vault automatically as your situation changes -- new role, new project, new priorities. You can also update any file manually at any time, upload documents, or chat with your context directly to add things the extraction didn't capture.

MCP (Model Context Protocol) is an open standard, co-developed by Anthropic, OpenAI, and Block, that lets AI tools pull structured data from external sources. When connected via MCP, Claude, Cursor, or any MCP-compatible agent pulls your context automatically at the start of each session.

MCP is the recommended way to use Unabyss, but not the only way. You can also use one of 20+ one-click exports -- pre-formatted context files ready to paste into a specific tool -- or create a custom export. All exports are Markdown files you can upload anywhere.

Mem0 is developer infrastructure -- an API for engineering teams building memory into AI products. Unabyss is for the users of AI tools, not the builders. No SDK, no API, no integration work -- connect your sources, and your context is ready to serve to Claude, Cursor, or any MCP tool, without a single line of code.

Yes -- this is a core feature, not a footnote. Your context is organised in three layers:

Identity -- public facts: name, role, company, location. Safe for any tool.

Profile -- professional history, goals, communication style. Shareable with most tools.

Mind -- private notes, reflections. Opt-in only, never shared by default.

When any app requests your context, you see exactly what's being requested and can approve, deny, or edit access per layer and per file.

Your context is encrypted at rest (AES-256) and in transit (TLS 1.3). EU users get EU-hosted data, GDPR-compliant from day one. Every app connection uses three-legged OAuth -- you authorise access on Unabyss, a scoped token is issued to the app. You can revoke any app's access instantly. Unabyss does not sell your data or use it to train AI models. See our Privacy Policy and Terms of Service for details.

All access tokens for that app are invalidated immediately. The app can no longer request your context. You can re-grant access at any time from your permissions dashboard.

Yes, at any time. Unabyss offers 20+ one-click exports -- pre-formatted context files optimised for specific tools -- as well as custom exports for any format you need. All exports are Markdown files you can upload directly into Claude, ChatGPT, Notion AI, or any other tool that accepts text input.

Unabyss cross-references facts across all your connected sources. If your LinkedIn says one thing and your Notion says another, Unabyss flags the conflict rather than arbitrarily picking one. This multi-source arbitrage is what makes the output more reliable than pulling from any single source alone.

You can review and edit your context vault at any time, but you don't have to. You can also update any file manually, upload documents, or chat with your context directly to add things the extraction didn't capture.

No. Unabyss works natively via MCP -- a single config file edit connects it to Claude Desktop, Cursor, or any MCP-compatible agent. No browser extension required. A Chrome extension is coming soon for users who prefer in-browser context injection.

Yes. They serve different purposes: platform memory records what AI learns from your conversations over time; Unabyss provides structured, intentional context from day one. Many users run both -- Unabyss handles "who I am and what I'm working on," platform memory handles "what we discussed last time."

Via MCP: Claude Desktop, Claude Code, Cursor, and any other MCP-compatible agent -- including OpenClaw.

Via exports: any AI tool that accepts text input -- ChatGPT, Gemini, Notion AI, and others. 20+ one-click exports or custom exports, all as Markdown files.

Via REST API: custom integrations for technical users.