GitHub Copilot app slash commands provide shortcuts for managing sessions, planning, and automating development workflows. Key commands include /plan for breaking down tasks before coding, /spar for pressure-testing architectural decisions, /autopilot for handing off multi-step implementation tasks, /rubber-duck for getting a second-model review of plans or code, /create-canvas for building interactive visualizations from chat, and /orchestrate for coordinating work across multiple repositories. Commands are triggered by typing / in the chat composer, which surfaces an autocomplete menu of context-aware options.
Table of contents
What are slash commands?Why use slash commands?Before you write code, make a /planLet Copilot play devil’s advocate with /spar/autopilot take the wheelTalk it through with /rubber-duckTurn conversations into interactive experiences with /create-canvasWhen one task becomes many, use /orchestrateYour next workflow starts with /Tags:Written byQuestions this post answers
What does the /spar slash command do in the GitHub Copilot app?
/spar has Copilot act as a devil's advocate, challenging your assumptions and pointing out risks or tradeoffs in your approach before you commit to a solution. You can use it to validate architecture choices, compare implementation options like REST vs GraphQL, review migration plans, or critique performance optimizations — essentially pressure-testing any technical decision. Developers weighing architecture trade-offs track tools like this on daily.dev.
What is the difference between /plan and /autopilot in the GitHub Copilot app?
/plan switches the session into Plan mode and helps you break down a task, identify files and dependencies, and outline an approach before writing code. /autopilot switches into Autopilot mode and takes over implementation — you give it a goal and it works through the steps, makes changes, and iterates, handling multi-step tasks like dependency updates or feature builds end to end. Teams deciding how to integrate AI coding agents into their workflow find comparisons like this on daily.dev.
What does /rubber-duck do in the GitHub Copilot app and how is it different from regular chat?
/rubber-duck uses a different model to independently review your work, surfacing blind spots, questioning assumptions, and catching issues the primary model may have missed. It is particularly useful for complex refactors, architectural decisions, and migration plans — any situation where a second opinion from a separate model adds value before you proceed. Developers building review checkpoints into AI-assisted workflows find relevant coverage on daily.dev.