MCP servers
Things I connect.
Documentation for a few Model Context Protocol prototypes I built for my own use. The hosted endpoints are currently offline; these pages remain as a technical archive.
AI-powered interface for Whoop fitness data. Recovery scores, sleep analysis, workout insights, and body metrics — in natural language.
Read the docs →Query and analyze Google Ads campaigns in plain English. Performance metrics, optimisation hints, and reports on demand.
Read the docs →Connect AI assistants to Meta advertising data. Facebook and Instagram ad performance, audiences, and creative insights.
Read the docs →What is MCP?
An open standard that lets AI assistants talk to real systems.
The Model Context Protocol is an open standard that lets assistants like Claude, ChatGPT, and Cursor connect securely to external data sources and tools. The examples below document local prototypes, not currently available hosted services.
- 01
Read the prototype notes
Each page records the intended tools and local setup.
- 02
Review the source pattern
Use the examples as historical implementation reference.
- 03
Verify before reuse
No hosted endpoint or maintained connection is promised here.