GitHub Copilot CLI

github.com · CLI Tool · Terminal

GitHub Copilot CLI is a CLI tool: GitHub Copilot coding agent in terminal with MCP extensibility. Its repository github/copilot-cli carries 11,244 stars and 26 contributors, and is licensed Other. The latest stable release is v1.0.93, published 7 Oct 2026 with binaries for linux, macos, windows. npm served 1,723,666 downloads of @github/copilot in the week we read. Copilot CLI shows every action before it runs, and nothing executes until you approve it.

Stars: 11,244; Verified 8 Oct 2026
Last commit: 7 Oct 2026; 0 days old when we read it; Verified 8 Oct 2026
License: Other; Verified 8 Oct 2026
Latest stable release: v1.0.93; published 7 Oct 2026; Verified 8 Oct 2026
Contributors: 26; Verified 8 Oct 2026
npm downloads: 1,723,666 a week; @github/copilot; Verified 8 Oct 2026
Repository: github/copilot-cli; Shell; Verified 8 Oct 2026
Release binaries: linux, macos, windows; Verified 8 Oct 2026

Visit GitHub Copilot CLI github/copilot-cli on GitHub GitHub Copilot CLI alternatives

What GitHub Copilot CLI is

Copilot CLI bills the agent from a dollar-denominated credit balance rather than per token: Pro carries $15 of GitHub AI Credits a month, Pro+ $70 and Max $200.

“Credits used when interacting with Copilot features. Chat, agent mode, code review, Copilot cloud agent, Copilot CLI, and Copilot Apps consume GitHub AI Credits. Free plan supports CLI and agent mode.”

github.com/features/copilot

What GitHub Copilot CLI costs

Copilot CLI is on the $0 Free plan; paid plans are Pro at $10 per user per month, Pro+ at $39 and Max at $100.

“Pro · For everyday coding with agents in GitHub Copilot. · $ 10 USD per user / month”

github.com/features/copilot

The Max plan is built for sustained agent work and carries $100 a month of GitHub AI Credits.

“GitHub Copilot Max is built for heavy Copilot usage, including sustained agent-driven workflows, and includes $100/month in GitHub AI Credits.”

github.com/features/copilot

Use GitHub Copilot CLI when

Work across code and GitHub state — issues, pull requests, repositories — in one session, because the agent ships with GitHub's MCP server enabled.

“- **MCP-powered extensibility:** Take advantage of the fact that the coding agent ships with GitHub's MCP server by default and supports custom MCP servers to extend capabilities.”

github.com/github/copilot-cli

Hold one very long session: at 95% of the token limit Copilot compacts the history in the background instead of stopping.

“Auto-compaction : When your conversation approaches 95% of the token limit, Copilot automatically compresses your history in the background without interrupting your workflow. This enables virtually infinite sessions.”

docs.github.com/en/copilot

Agree the scope before any code is written: plan mode asks clarifying questions and produces a plan first.

“In plan mode, Copilot analyzes your request, asks clarifying questions to understand scope and requirements, and builds a plan before writing any code.”

docs.github.com/en/copilot

Not the right fit for

A developer with no GitHub Copilot subscription who wants an agent on day one should pick OpenCode or Crush; Copilot CLI needs an active Copilot plan.

“- An **active Copilot subscription**. See [Copilot plans](https://github.com/features/copilot/plans?ref_cta=Copilot+plans+signup&ref_loc=install-copilot-cli&ref_page=docs).”

github.com/github/copilot-cli

What GitHub Copilot CLI is bad at

Your employer can switch the tool off: when Copilot comes from an organization, an administrator must enable the Copilot CLI policy first.

“GitHub Copilot CLI is available with all Copilot plans. If you receive Copilot from an organization, the Copilot CLI policy must be enabled in the organization's settings.”

docs.github.com/en/copilot

You cannot bring an arbitrary model: one that lacks tool calling or streaming makes the CLI return an error, and GitHub asks for a 128k context window.

“Models used with Copilot CLI must support tool calling (function calling) and streaming . If the model does not support these capabilities, Copilot CLI will return an error. For best results, the model should have a context window of at least 128k tokens.”

docs.github.com/en/copilot

Model access fails in the field: a 403 on listing models is open with 30 comments, and sporadic policy blocking on model retrieval is open with 29.

“Failed to list models: 403 : unauthorized: not authorized to use this Copilot feature”

github.com/github/copilot-cli

Security and data handling

Copilot CLI shows every action before it runs, and nothing executes until you approve it.

“- **Full control:** Preview every action before execution — nothing happens without your explicit approval.”

github.com/github/copilot-cli

Technical details

Category: CLI Tool · Terminal
Repository: github/copilot-cli
Default branch: main
Primary language: Shell
Repository created: 6 Jan 2023
Newest commit on the default branch: 7 Oct 2026
Package formats: msi, tar.gz, zip
Architectures: arm64, x64, x86
Registries checked: npm:@github/copilot,typescript-language-server,github-copilot-cli, homebrew:copilot-cli,github-copilot-cli, vscode_marketplace:search(GitHub Copilot CLI)
Website check: https://github.com/github/copilot-cli answered HTTP 200

Alternatives to GitHub Copilot CLI

Claude CodeCommand-line AI coding agent that understands your codebase and automates tasks via natural languageCompare
AiderAI pair programming in your terminal with clean commitsCompare
Codex CLIOpenAI's lightweight terminal chat interface for code operationsCompare
GooseLocal on-machine AI agent that allows any LLM and MCP servers as extensionsCompare
MyCoder.aiOpen source AI coding assistant with Git/GitHub integration and parallel executionCompare

All GitHub Copilot CLI alternatives, ranked →

How we check this page

Machine facts. Stars, licence, the newest commit on the default branch, the latest stable release and download counts are read straight from the GitHub API and from npm, PyPI, Homebrew and the VS Code Marketplace. Each one shows the day it was last confirmed, not the day this page was built. A failed fetch keeps the old value and its old date rather than overwriting it.
Judgments. Every claim about what a tool suits, what it is bad at and what it costs is quoted from a page we read, and that page is linked beside the claim. If there is no link, we did not write the claim.
What we do not publish. No score out of five and no star rating. Every other claim here traces to a source, and a number we could not justify would undermine the ones that can.
This tool. Every claim carries the URL it came from and a verbatim quote from that URL, read on 2026-08-19.