copilot-cli

Delegate multi-step repository work to GitHub Copilot CLI non-interactively.

5|Updated Apr 23, 2026
One-click install
npx skills add https://github.com/amirkiarafiei/subagent-cli-skills --skill copilot-cli-amirkiarafiei
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: copilot-cli
Source: https://github.com/amirkiarafiei/subagent-cli-skills/tree/main/skills/copilot-cli
Command: npx skills add https://github.com/amirkiarafiei/subagent-cli-skills --skill copilot-cli-amirkiarafiei

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

GitHub Copilot CLI enables your orchestrator to offload large, multi-step engineering work into a separate headless execution pass, reducing your main agent’s context load and token costs.

Core Features & Use Cases

  • Specialist agent delegation: Run Copilot with focused agents like explore, general-purpose, task, research, or code-review to match the job type.
  • Context handoff for isolated runs: Provide the full goal, prior decisions, scope, constraints, and verification so the isolated session stays aligned.
  • Non-interactive, programmatic execution: Use flags like non-interactive prompt mode, silent output, model selection, effort level, and auto-tool approval for reliable automation.
  • Common use cases: deep repo exploration, broad refactors across many files, multi-file code generation/edits, and review-style passes for security and correctness.

Quick Start

Ask your orchestrator to delegate the work to Copilot CLI using a single prompt that includes the original goal, prior decisions, file scope, constraints, the exact verification command to run, and the expected output format.

Frequently Asked Questions about copilot-cli

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I delegate large multi-file refactoring tasks to Copilot CLI?

You can delegate large multi-file refactoring to Copilot CLI by running it non-interactively with a single prompt that includes the full goal, prior decisions, scope, constraints, and verification commands.

What context do I need to pass for isolated Copilot CLI code generation?

For isolated Copilot CLI code generation, you must pass the complete goal, prior decisions, file scope, constraints, and the exact verification command to keep the headless session aligned with the main thread.

Can I run Copilot CLI non-interactively for automated code review?

Yes, you can run Copilot CLI non-interactively for automated code review by using silent output, model selection, effort level flags, and auto-tool approval for reliable headless automation.

Does Copilot CLI support specialist agents for repo exploration?

Yes, Copilot CLI supports focused specialist agents like explore, general-purpose, task, research, and code-review to match specific job types such as deep repo exploration or broad refactors.

Why use Copilot CLI for headless automation instead of my main agent?

Using Copilot CLI for headless automation offloads large, multi-step engineering work into a separate execution pass, which reduces your main agent's context load and token costs.

What's the best way to automate broad codebase refactoring with Copilot CLI?

The best way to automate broad codebase refactoring is delegating the work to Copilot CLI using non-interactive prompt mode with specified model and effort levels, including complete scope and verification details.