MCP Notes
AI agents are powerful — but they forget everything between conversations. xNotePadAI is built around MCP from the ground up — 13 tools, agent sandboxing, zero-knowledge encryption, and mandatory diff approval. Your notes become persistent, private memory that any AI can use safely.
13 MCP Tools
Read, write, search, tag, link, merge, archive, create, list, ask, organise, and export — all available to any MCP-compatible AI agent.
Bearer Token Auth
Secure access via Bearer token authentication. You control which agents connect and can revoke access at any time.
Works with Claude
Connect Claude Desktop and say "save this research to my notes" or "what did I write about auth architecture?" — instant access to your knowledge.
Works with Cursor
Your coding agent can reference project decisions, architecture notes, and meeting outcomes stored in xNotePadAI while it writes code.
Diff Approval
When AI agents modify your notes, every change requires your approval via a visible diff. No silent edits, ever — even from MCP.
Semantic Search
AI agents can search your notes by meaning using vector embeddings. "Find notes about performance optimisation" works even if you never used that exact phrase.
Why xNotePadAI's MCP is different
Several note apps now offer MCP connections. xNotePadAI differentiates with: agent sandboxing (agents can only modify notes they created), mandatory diff approval, zero-knowledge encryption, confidential section exclusion, and tool annotations that tell clients which operations are destructive.
Connect your AI agent in under 2 minutes.
Frequently Asked Questions
What is MCP?
Model Context Protocol is an open standard that lets AI agents interact with external tools and data sources. It's how Claude, Cursor, and other AI systems access APIs — think of it as USB for AI.
Which AI agents are supported?
Any agent that supports the Model Context Protocol — currently Claude Desktop, Cursor, Kiro, and a growing list of MCP-compatible tools. The protocol is open, so new agents gain access automatically.
Is my data safe if AI agents can access it?
Access is secured with Bearer token authentication that you generate and control. You can revoke access at any time. All write operations require your explicit approval via diff. AI agents cannot silently modify or delete your notes.
What can AI agents actually do with my notes?
They can read notes, search by content or tag, create new notes, suggest edits (with your approval), tag and link notes, merge related content, and archive old notes. The 13 tools cover the full lifecycle of knowledge management.