# GitHub Copilot CLI

56 Hacker News commenters recommended GitHub Copilot CLI and 37 criticized it, November 2025 to November 2026. It was recommended in 12 of 13 months; owners who said how long report a median of 0.25 years (3 owners).

**Verdict:** People recommend it as a capable terminal coding agent, often close to Claude Code, and it works well with a Copilot plan. Critics mention weaker harness, slowness and memory leaks.

## Why people recommend it

- Improving quickly, team responsive to feedback (3 comments): "The team's been working really hard to improve the harness, and we're taking as much community feedback as we can get!" ([source](https://news.ycombinator.com/item?id=46748608)) "The Copilot CLI team has been making great strides towards improving our agentic harness!" ([source](https://news.ycombinator.com/item?id=46748631)) "which has got much better over the past few months" ([source](https://news.ycombinator.com/item?id=48241209))
- Iterates until the task is done (2 comments): "I tried the CLI, and for some reason it was much better. I explain carefully what I want, and it iterates until it's done, quickly, on cheap models." ([source](https://news.ycombinator.com/item?id=46962476)) "Copilot will keep working until it fulfills your request, no matter how many tokens it uses." ([source](https://news.ycombinator.com/item?id=46972906))
- Close to Claude Code capability (2 comments): "It's pretty good, for my purposes at work it's probably as good as Claude Code." ([source](https://news.ycombinator.com/item?id=46864978)) "It's not bad, I would say it's about 80% as capable as Claude Code" ([source](https://news.ycombinator.com/item?id=47927085))
- Flexible across many models (3 comments): "The mentioned ability to easily call on various models in the same prompt" ([source](https://news.ycombinator.com/item?id=48760121)) "Teams where I work can use Claude Code, Codex, Cursor, and Copilot CLI." ([source](https://news.ycombinator.com/item?id=47420814)) "Been having a ton of fun with copilot cli directed to local qwen 3.6" ([source](https://news.ycombinator.com/item?id=47857723))

## What people warn about

- Can bypass read-only command restrictions (4 comments): "The env command is part of a hard-coded read-only command list stored in the source code." ([source](https://news.ycombinator.com/item?id=47190916)) "It shouldn't be able to just do `env sh` and run whatever it wants without approval." ([source](https://news.ycombinator.com/item?id=47190770)) "Why even have any user confirmations at all if the LLM can just bypass them at will?" ([source](https://news.ycombinator.com/item?id=47198415))
- Weaker harness than Claude Code (3 comments): "Copilot CLI is worse than Claude Code, but not by a huge amount." ([source](https://news.ycombinator.com/item?id=47391754)) "Copilot's harness and the underlying Claude models are not well-matched compared to the vertically-integrated Claude Code harness." ([source](https://news.ycombinator.com/item?id=47377210)) "Claude Code has something special that Microsoft doesn't have in Github Copilot CLI" ([source](https://news.ycombinator.com/item?id=46857151))
- Recent changes hurt usability and cost (3 comments): "found it now provided a degraded experience because Microsoft unanimously decided to change some part of it without my consent" ([source](https://news.ycombinator.com/item?id=49206126)) "After the recent changes to usage, I've spent an annoyingly long number of hours trying to get this to work." ([source](https://news.ycombinator.com/item?id=48559209)) "copilot cli became even more unusable than claude pro subscription around January/February" ([source](https://news.ycombinator.com/item?id=48355855))

## Alternatives hackers recommend

- [Cursor CLI](https://hackerapproved.com/p/cursor-cli.md): 17 recommend

## Where to get it

- [Visit the official site](https://github.com/features/copilot)

## Cite as

Hacker Approved, "GitHub Copilot CLI", https://hackerapproved.com/p/github-copilot-cli/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).
