Skip to content

Image/Screenshot Paste Support in Copilot CLI #3370

@LanePalmer

Description

@LanePalmer

Describe the feature or problem you'd like to solve

I can't paste screenshots or images directly into the Copilot CLI chat. When troubleshooting visual issues (error dialogs, UI bugs, etc.), I have to describe what I'm seeing in text instead of just showing it — which slows things down and loses context. Today I was troubleshooting a persistent macOS popup and couldn't simply paste a screenshot for Copilot to see, which would have made the whole interaction faster and more accurate.

Proposed solution

One of the biggest friction points in my daily Copilot CLI workflow is not being able to paste screenshots or images directly into the chat.

Real example from today: I had a persistent macOS popup that I was trying to troubleshoot with Copilot. Instead of being able to paste a screenshot so Copilot could see exactly what I was seeing, I had to manually describe it in text — which slows things down and introduces room for miscommunication.

For anyone using Copilot CLI for real-world troubleshooting, debugging, or workflows that involve UI elements, error messages, or visual context, the inability to share images is a meaningful gap compared to chat-based AI tools.

Supporting image paste (even just screenshots via Cmd+V) would make Copilot CLI dramatically more useful for day-to-day work and bring it in line with the visual context capabilities users expect from modern AI assistants.

Example prompts or workflows

  1. Troubleshooting UI issues: Paste a screenshot of an error dialog or popup and ask "what is this and how do I fix it?" — instead of having to describe it manually.

  2. Code review from screenshots: Paste a screenshot of code or a diff from another tool and ask Copilot to review or explain it.

  3. Debugging terminal output: Paste a screenshot of a terminal error and ask "what's going wrong here?" — useful when the error can't be easily copied as text.

  4. Design/UI feedback: Paste a screenshot of a UI and ask Copilot to help write code that replicates or improves it.

  5. Diagrams and architecture: Paste a system architecture diagram and ask Copilot to help implement or explain components of it.

Additional context

Most modern AI chat interfaces (ChatGPT, Claude, Gemini) already support image input, so users have come to expect it as a baseline capability. Bringing this to Copilot CLI would close a meaningful gap and make it the obvious choice for terminal-based AI workflows — especially for power users who live in the CLI and don't want to switch to a browser-based tool just to share visual context. Even a simple Cmd+V to paste a screenshot would go a long way.

Metadata

Metadata

Assignees

No one assigned

    Labels

    area:input-keyboardKeyboard shortcuts, keybindings, copy/paste, clipboard, mouse, and text inputarea:modelsModel selection, availability, switching, rate limits, and model-specific behavior
    No fields configured for Feature.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions