skillpack.co
All skills

GitHub Copilot CLI

active

GitHub's official terminal coding agent. GA Feb 25, 2026. Multi-model (Opus 4.6, Sonnet 4.6, GPT-5.3-Codex, Gemini 3 Pro). Enterprise Agent Control Plane. 15M Copilot subscriber distribution. 9.4K stars.

Connector
Orchestrator
Complexity
clicoding

10/100

Trust

1.1K+

Stars

3

Evidence

Videos

Reviews, tutorials, and comparisons from the community.

The ultimate guide to the GitHub Copilot CLI | Full demo | GitHub Checkout

GitHub·2025-10-02

Power agentic workflows in your terminal with GitHub Copilot CLI

GitHub·2025-11-02

Demo: Using GitHub Copilot CLI and yolo mode

GitHub·2026-01-22

Repo health

10/100

4mo ago

Last push

63

Open issues

89

Forks

5

Contributors

This repository is archived and no longer maintained.

Editorial verdict

#5 coding CLI — distribution is the killer advantage (15M paid Copilot subscribers). Multi-model approach and Enterprise Agent Control Plane are unique. But CVE-2026-29783 (arbitrary code execution 2 days after GA) and no published benchmark scores are serious concerns.

Source

Public evidence

strong2026-02
PromptArmor: CVE-2026-29783 — arbitrary code execution in Copilot CLI

A crafted env curl command slips past the validator, downloads a payload from an attacker URL, and pipes it to sh. No confirmation dialog, no approval. Hit 2 days after GA (Feb 27, 2026).

Security research disclosure, CVE assignedPromptArmor (independent security research firm)
moderateSelf-reported2026-02
GitHub Copilot CLI GA launch — multi-model terminal agent

GA launch with multi-model support (Claude, GPT, Gemini), Plan mode, Autopilot, Agent Skills. Enterprise Agent Control Plane for org admin controls. 15M subscriber distribution.

Official GitHub blog, 24 HN pointsGitHub (official)
moderate2026-02
Morph 15-agent comparison: Copilot CLI classified Tier 2

Morph's independent tier classification puts Copilot CLI in Tier 2. Distribution advantage (15M subscribers) is acknowledged but capability is unproven without benchmark scores.

Independent 15-agent comparison studyMorph Engineering Team (independent AI infrastructure company)

How does this compare?

See side-by-side metrics against other skills in the same category.

COMPARE SKILLS →

Where it wins

15M paid Copilot subscribers get it automatically — largest distribution in category

Multi-model: Claude Opus 4.6, Sonnet 4.6, GPT-5.3-Codex, Gemini 3 Pro — users pick per task

Enterprise Agent Control Plane — most mature enterprise admin controls in category

Feature set: Plan mode, Autopilot, /research deep research, Agent Skills

GA since Feb 25, 2026 — shipping as a full product, not beta

Where to be skeptical

CVE-2026-29783 hit 2 days after GA — arbitrary code execution via shell expansion, 'no confirmation dialog, no approval' (PromptArmor)

No published SWE-bench Pro or Terminal-Bench scores

Requires Copilot subscription — not free standalone

Low community signal: 24 HN points on launch, 9.4K stars (lowest among serious contenders)

Quality of the agent itself is unproven — distribution ≠ capability

Ranking in categories

Know a better alternative?

Submit evidence and we'll run the full pipeline.

SUBMIT →

Similar skills

Raw GitHub source

GitHub README peek

Constrained peek so you can sanity-check the source material without leaving the site.

GitHub Copilot in the CLI has been deprecated on October 25, 2025 in favor of GitHub Copilot CLI, an agentic assistant that brings AI-powered coding assistance directly to your command line, enabling you to build, debug, and understand code through natural language conversations. Powered by the same agentic harness as GitHub's Copilot coding agent, it provides intelligent assistance while staying deeply integrated with your GitHub workflow. We suggest migrating to the GitHub Copilot CLI, which can explain and suggest shell commands and much more!

GitHub Copilot in the CLI

GitHub Copilot in the CLI is an extension for GitHub CLI which provides a chat-like interface in the terminal that allows you to ask questions about the command line. You can ask Copilot in the CLI to suggest a command for your use case with gh copilot suggest, or to explain a command you're curious about with gh copilot explain.

For enabling and using, see "Using GitHub Copilot in the command line".

For use cases and limitations, see "Responsible use of GitHub Copilot in the CLI".

For what data is collected, used, and shared, see "Privacy Policies".

For help troubleshooting connectivity, see "Troubleshooting network errors for GitHub Copilot".

Quickstart

[!NOTE] To use and install GitHub Copilot in the CLI, you must have an active GitHub Copilot subscription, have GitHub CLI installed, and authenticate using the GitHub CLI OAuth app.

Classic and fine-grained PATs are currently unsupported and might require clearing the GITHUB_TOKEN and GH_TOKEN environment variables.

[!IMPORTANT] GitHub Copilot in the CLI does not currently have plans to support 32-bit Android distributions.

For more information, see https://github.com/github/gh-copilot/issues/122.

  1. Authenticate with GitHub CLI OAuth app
    gh auth login --web
    
  2. Install / upgrade extension
    gh extension install github/gh-copilot --force
    
  3. Suggest a command
    gh copilot suggest "Install and configure git lfs"
    
  4. Explain a command
    gh copilot explain 'git lfs migrate import --everything --include="*.gz,*.png,*.jar"'
    

Usage

$ gh copilot --help
Your AI command line copilot.

Usage:
  copilot [command]

Examples:

$ gh copilot suggest "Install git"
$ gh copilot explain "traceroute github.com"


Available Commands:
  alias       Generate shell-specific aliases for convenience
  config      Configure options
  explain     Explain a command
  suggest     Suggest a command

Flags:
  -h, --help              help for copilot
      --hostname string   The GitHub host to use for authentication
  -v, --version           version for copilot

Use "copilot [command] --help" for more information about a command.
Multi-account notes

Multi-account users can use --hostname flag or GH_HOST environment variable to specify the GitHub hostname for commands that would otherwise assume the github.com host.

Set up optional helpers

Is gh copilot suggest ... too many keystrokes? Do you want support for executing suggestions and keeping them in history?

v1.0.0 introduces gh copilot alias, which generates shell configuration for ghcs and ghce aliases that wrap gh copilot suggest and gh copilot explain. These aliases provide faster invocation and support executing suggested commands if applicable. Executed suggestions are added to your shell history for reuse later.

$ ghcs print "Hello world"

Welcome to GitHub Copilot in the CLI!
version 1.0.0 (2024-03-21)

I'm powered by AI, so surprises and mistakes are possible. Make sure to verify any generated code or suggestions, and share feedback so that we can learn and improve. For more information, see https://gh.io/gh-copilot-transparency

Suggestion:

  echo "Hello world"

? Select an option
> Execute command

? Are you sure you want to execute the suggested command?
> Yes

Hello world

To setup these optional helpers, see below for setup instructions for supported shells:

  • bash
  • powershell
  • zsh

For more information, run gh copilot alias --help.

For changing the execute command confirmation behavior, run gh copilot config to change Default value for confirming command execution.

Bash
echo 'eval "$(gh copilot alias -- bash)"' >> ~/.bashrc
PowerShell

[!NOTE] PowerShell users might need to update the file containing gh-copilot helpers after updating the extension.

$GH_COPILOT_PROFILE = Join-Path -Path $(Split-Path -Path $PROFILE -Parent) -ChildPath "gh-copilot.ps1"
gh copilot alias -- pwsh | Out-File ( New-Item -Path $GH_COPILOT_PROFILE -Force )
echo ". `"$GH_COPILOT_PROFILE`"" >> $PROFILE
Zsh
echo 'eval "$(gh copilot alias -- zsh)"' >> ~/.zshrc

Metrics sent to GitHub

GitHub Copilot in the CLI sends certain metrics to our analytics system. We want you to understand what is being sent and why it's important to our ability to continue to improve the product and provide you with a better experience

View on GitHub →