Bring real code intelligence to your terminal. The new LSP Setup skill for GitHub Copilot CLI adds semantic context (like definitions, references, and types) across 14 languages. Here’s how to get started 👇 github.blog/ai-and-ml/gith…

@github why would i ever use this instead of codex or claude code 🤔

@github LSP in the terminal is nice, but wake me up when it stops hallucinating the 14 languages' docs. 🤔

@github I was a great fan of GH Copilot, until it's loud change around a month ago

@github Nice — semantic LSP context in the terminal is exactly what makes agentic workflows feel usable. Anything that cuts token burn while surfacing defs/types is a win.

@github Tks. Good to know.

@github lsp cold-init is the quiet killer. dart analysis server taught me this the hard way. first find-references call returns nothing until the workspace scan finishes. 40+ seconds on a big project.

@github definitions, references, and types. everything except the will to refactor

@github Developers are spoilt for choice

@github Semantic setup sounds small until the agent edits the wrong symbol in a language it only half understands. I’d rather give the tool compiler-grade context than hope grep found the right thing.

Interesting but... We used to run a lot of GitHub Copilot at my company, but after the new billing system rolled out, we basically moved fully to Claude and Codex. Not because Copilot was bad. But because Claude/Codex became more economical and effective over time. Before that, developers used Copilot as a parallel track a way to get a broader perspective than just 1–2 models from the same vendor

Our delivery comes with a genius. Go learns what you like, builds your cart in seconds, and gets it all to your door in as fast as 15 minutes.

@github this seems to be interesting 😎

@github lol who uses copilot in 2026 😂

@github 👏👏👏

@github Best code intilligence