A practical MCP surface
Exposes 69 product capabilities across tasks, habits, notes, and analytics so agents can retrieve information and take useful actions through a standardized interface.
Private project · AI agents
An AI integration layer for a personal productivity product. It combines provider abstractions, tool calling, and an MCP server so an assistant can work with the product’s data through clear, typed capabilities.
Visit hubly.devWhat it does
Exposes 69 product capabilities across tasks, habits, notes, and analytics so agents can retrieve information and take useful actions through a standardized interface.
Supports both OpenAI function calling and Claude tool use behind one backend, allowing the product to evolve without coupling its AI experience to one provider.
Connects the agent layer to Supabase-backed product data and analytics, turning natural-language requests into product-aware workflows.
Built with
Want to talk through the thinking behind it?
I'm always interested in useful product and platform conversations.