Unabyss
Shared memory layer that feeds Gmail, Slack and Notion context to AI clients
- Pricing
- Free trial
- Platforms
- Web, API
- Runs
- Remotely
- Version
- 1.0.0
- Listing updated
- 1 day ago
About Unabyss
Unabyss is an MCP server for knowledge and memory. It runs as a hosted remote server, so there is nothing to install locally.
Most AI assistants start every chat blank and keep whatever they learn locked inside one product. Unabyss tries to fix that by pulling context from work apps, including Gmail, Slack, Notion, Calendar, GitHub and meeting notes, and structuring it into one memory that any MCP client can read.
The site says the context is tagged by topic, sensitivity and source, and only the relevant slice is returned for a query. It stays connected to the sources so it updates itself, and it also takes in what you do inside Claude, ChatGPT or Cursor. The service lists a Claude connector, a ChatGPT plugin and a one-click Cursor install, and the context can be exported.
On security, the site mentions AES-256 at rest, TLS 1.3, OAuth scoped per app and source, EU-hosted data for EU users and no training on user data; SOC 2 Type II is shown as in progress.
Install
Run in a terminal
claude mcp add --transport http unabyss https://mcp.unabyss.com/Add to Settings → Connectors → Add custom connector
https://mcp.unabyss.com/Add to ~/.cursor/mcp.json
{
"mcpServers": {
"unabyss": {
"url": "https://mcp.unabyss.com/"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"unabyss": {
"type": "http",
"url": "https://mcp.unabyss.com/"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"unabyss": {
"serverUrl": "https://mcp.unabyss.com/"
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"unabyss": {
"type": "http",
"url": "https://mcp.unabyss.com/"
}
}
}What it does
- Cross-tool memory
- One context shared between Claude, ChatGPT, Cursor and other MCP clients.
- Live source sync
- Stays connected to Gmail, Slack, Notion, Calendar, GitHub and 20+ more sources.
- Structured context
- Tags content by topic, sensitivity and source and returns only the relevant part.
- Scoped OAuth access
- Access is granted per app and per source and can be revoked.
- Exportable context
- Context files can be exported and moved between models.
Good for
- Stop re-explaining your role and projects to each AI tool
- Let Cursor see decisions made in Slack or Notion
- Carry context from Claude to ChatGPT
- Give a team a shared AI context
Pricing
Free trial, from $15 a month
7-day free trial on every plan; Pro $15/mo, Max $89/mo, Team $49/seat/mo (3-seat minimum); annual billing saves up to 20%.
Prices change. Check unabyss.com before you buy.
Unabyss FAQ
What does it cost?
Every plan has a 7-day free trial. Pro is $15/month, Max is $89/month and Team is $49 per seat per month with a 3-seat minimum; there is no pay-as-you-go option.
How is it different from built-in AI memory?
Built-in memory only works inside the tool that made it, while Unabyss sits above the tools and pulls from your work apps.
How long does setup take?
The site says under 90 seconds: connect sources, then add Unabyss to Claude, ChatGPT or Cursor.
Is my data used for training?
The site says it does not sell data or use it to train AI models.
Is Unabyss free?
Unabyss is a paid product with a free trial. Paid plans start at $15 a month.
How do I add Unabyss to Claude Code?
Run this command in your terminal: claude mcp add --transport http unabyss https://mcp.unabyss.com/
What are the best alternatives to Unabyss?
The closest alternatives to Unabyss are Codebase Memory, XERJ, Ha MCP, Claude Code Ultimate Guide and Nautilus Compass Memory.
Alternatives to Unabyss

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ha MCP
MCP serverMCP server for managing and automating Home Assistant from AI assistants
Runs locally- 8 tools
- Open source
- 7.9K GitHub stars
- Productivity

Claude Code Ultimate Guide
MCP serverMCP server for searching the Claude Code Ultimate Guide from any MCP client
Runs locally- Open source
- 6.5K GitHub stars
- AI models and LLM tooling

Nautilus Compass Memory
MCP serverLocal-first long-term memory layer for AI agents, served over MCP
Not checked yet- 5 tools
- Remote
- Open source
- 1.3K GitHub stars
- Knowledge and memory

Concord MCP
MCP serverLocal-first MCP server for messaging and shared work-state between coding agents
Runs locally- Freemium
- 1.3K GitHub stars
- Developer tools