8 docs tagged with "MCP"

How to engineer agent-grade tools — CLIs, MCP servers, and the contracts that let an agent operate a capability without guessing.

Reusable concept note for Agent-Readable Conversion.

Decision checklist and adoption radar for MCP tool selection across agent teams

Choose the AI tools that bring value to a workflow — which model, which CLI, which MCP server — and when each one is worth adopting.

Reusable concept note for MCP Server Catalog.

How to build a Model Context Protocol server — architecture, the three primitives, transports, and the token economics that decide whether an agent can afford your tools.

Choose which MCP servers give agents the most value per token — adoption radar, team profiles, and the server catalogue.

How AI agents access external tools, data, and resources — the universal adapter between models and the world