MeridiaanNotes on working with Agents
Practical writing on context, memory and shared knowledge: how teams and their AI Agents work on the same project without losing the thread.
What is an MCP Server?
The Model Context Protocol is quickly becoming the standard way for AI agents to talk to external tools and data sources. Here is what it is, how it works, and why it matters.
Why Your AI Agent Keeps Losing Context
Every new session, you explain the same things again. The architecture, the constraints, the decisions already made. There is a better way.
The Knowledge Base Your AI Agent Can Actually Use
Not all documentation is useful to an agent. Here is what makes a knowledge base queryable, relevant, and worth maintaining.
Where Every MCP Client Keeps Its Config
The protocol is standardised. The config file is not. A field guide to mcpServers vs servers vs mcp, project scope vs global scope, and the one file more than one vendor agreed to read.
Start for free, no credit card needed.
Set up in minutes. Scale when your team grows.