Enables Large Language Models (LLMs) to save and retrieve sticky notes directly within a web browser.
This tool serves as a simple demonstration of how Large Language Models (LLMs) can interact with a web browser to manage temporary information. By defining specific functions like `saveNote` and `readNote` via `window.mcp`, it allows compatible LLM clients (such as Claude Desktop, Glama.ai, or LobeHub) to programmatically save and read 'sticky notes.' This facilitates simple data persistence within a browser environment, showcasing practical browser automation for AI agents.
Key Features
01Auto-detectable and invokable by any MCP-compatible client.
02Allows LLMs to save notes in the browser.
03Allows LLMs to read notes from the browser.
04Exposes `saveNote` and `readNote` tools via `window.mcp` for LLM interaction.
050 GitHub stars
Use Cases
01Demonstrating browser-based data persistence for LLMs.
02Enabling LLMs to manage short-term information within a browser session.
03Providing a basic example of MCP tool definition and integration for developers.
