
MCP and skills: measure and reduce tokens without losing context
A reproducible MCP lab measures local token counts, tests filters and projections, and checks that shorter results preserve the evidence needed.
Blog / Matteo Vincenti
I write about AI in everyday work, software development and how teams work together. I start with business problems and explore the technical decisions.
3 articles in Context engineering
Topic: Context engineering Clear topic

A reproducible MCP lab measures local token counts, tests filters and projections, and checks that shorter results preserve the evidence needed.

Decide when a project map is enough and when to pin revisions. A complete submodule setup with manifests, agent instructions and a runnable Git example.

Organise context, sources and handoffs for AI agents. A real Notion knowledge path, from METR findings to reuse and ongoing maintenance.
Contact