Knowledge

A from-first-principles walkthrough of the read-plan-edit-verify loop that sits under every coding agent, built live with no framework.

A step-by-step guide to building, testing and shipping a Model Context Protocol server that exposes your own tools to an agent.

AI pair programming in your terminal. Works directly against your git repo and commits each change, so every edit is reviewable and revertible.

Evals, sandboxing and the unglamorous engineering that separates a demo from something you would let near a production repo.

What MCP is, what problem it solves, and how to write a server that a coding agent can actually use.

How to build an evaluation harness before you build the agent, so you can tell whether a change made things better or just different.

Context selection, chunking and verification strategies for making an agent safe to point at a million-line repository.

An open platform for software development agents, agents that can modify code, run commands, browse the web and call APIs.

The leading open-source AI code assistant. Connect any model to VS Code and JetBrains, and keep the whole loop self-hosted.

Reference MCP server implementations, filesystem, git, fetch and more. The best place to read real MCP code before writing your own.