Exposing local notes tools through MCP
Expose existing Go notes tools through a local MCP server, inspect discovery and calls, and keep file creation out of the protocol interface.
I write about language models, agents, and the code that makes them useful. These notes follow what I’m learning, with working examples and explanations of what happens underneath.
For the foundations, read how generative AI works. Next, explore agents and their protocols, or go straight to building a small model harness in Go.
Expose existing Go notes tools through a local MCP server, inspect discovery and calls, and keep file creation out of the protocol interface.
Build a small evaluation set for a Go model application and distinguish correctness checks from judgments about answer quality.
Build a read-only API tool with application-owned credentials, fixed endpoints, and resource authorization.
Build local notes tools in Go, validate every model request, and keep file creation behind separate human approval.
Run a small model locally, inspect its requests and responses, understand JSONL, and build a Go harness with conversation history and a restricted tool.
What agents do, how skills help them, and where A2A, MCP, AG-UI and ACP fit, with practical examples.
How generative AI learns and produces content, what changed with modern models, and when simpler software is a better choice.
How language models predict text, learn from data, and run on hardware, with diagrams and references.