GitHub Copilot CLI Now Understands Code Like Your Editor Does

If you've ever watched GitHub Copilot CLI unzip a JAR file, grep through raw bytecode, and try to piece together a method signature from the results, you already know the problem. The agent is resourceful, but without a language server, that's the best it can do. GitHub just shipped a fix: the LSP Setup skill, a new agent skill that wires Copilot CLI into real Language Server Protocol (LSP) servers for 14 languages.

The gap it fills

The Language Server Protocol (LSP) is the standard that powers go-to-definition, find references, and type resolution in editors like VS Code, and it works just as well in the terminal. Until now, Copilot CLI had no way to tap into it. Without an LSP server, the agent reverse-engineered API information through text search and binary extraction. For Python it would cat files inside site-packages; for TypeScript it walked node_modules. These heuristic approaches often missed crucial details like generics or overloads and couldn't interpret compiled bytecode effectively.

The structural fix is straightforward: when the agent sends a textDocument/definition request for a symbol, the language server returns the exact source location, fully resolved type, and signature. No guessing, no pattern-matching over raw text.

What an agent skill actually is

An agent skill is a reusable instruction set that extends what an AI coding agent can do. Skills are defined in Markdown files with YAML frontmatter and follow a standard structure: trigger descriptions, step-by-step workflows, reference data, and behavioral constraints. Think of it as a plugin that teaches the agent a new procedure, not just new facts. The LSP Setup skill uses this structure to guide the agent through a multi-step installation process, detecting the operating system, choosing the right package manager, writing valid configuration, and verifying the result.