VS Code Copilot

Set up CE.SDK with VS Code Copilot

GitHub Copilot in VS Code is an in-editor AI agent. Ten minutes of setup turns it into a CE.SDK expert that takes you from an idea like “a photo editor with filters” to a working editor running in your browser.

Before you start

You don’t need to be a developer to follow this guide. You need two things on your machine.

  • VS Code with the GitHub Copilot extension, signed in.
  • Node.js 20 or newer. It provides the npx command used below and runs the editor you build.

From setup to working editor in four steps

Step 01

Install the CE.SDK skills

Open a terminal in your project folder (in VS Code: Terminal, then New Terminal) and run one command. It installs the CE.SDK Agent Skills: the CE.SDK docs and starter kits for 10 web frameworks. GitHub ships its own installer for Copilot skills, which records the version and provenance in the skill frontmatter. It needs GitHub CLI 2.90 or newer.

Install the skills for Copilot
gh skill install imgly/agent-skills

The vendor-neutral Skills CLI works too, and installs into .agents/skills/ beside your project. Add -g to install once for every project on your machine.

Or use the Skills CLI
npx skills add imgly/agent-skills -a github-copilot
Step 02

Connect the live docs

The MCP server gives Copilot real-time search over the current CE.SDK documentation, so answers stay accurate between releases. Create a file named mcp.json inside the .vscode folder at your project root, paste this in, and reload VS Code.

.vscode/mcp.json
{
  "servers": {
    "imgly_docs": {
      "type": "http",
      "url": "https://mcp.img.ly/mcp"
    }
  }
}
Step 03

Verify it works

In Agent mode, check the tools picker for the imgly_docs tools, then ask a docs question and confirm the answer cites CE.SDK documentation.

Ask a test question
How do I add background removal to a CE.SDK editor?
Step 04

Describe what you want to build

Open Copilot Chat, switch to Agent mode, and paste a product idea. Copilot scaffolds the project, installs CE.SDK, wires up the features, and prints the command to run it (usually npm run dev). Run it and your editor opens in the browser.

Your first build prompt
Build a photo editor web app with CE.SDK: filters and adjustments, text overlays with our brand fonts, one-click background removal, and PNG export sized for social media.

What you can do

Three skills turn Copilot into a CE.SDK specialist, with live docs behind every answer. Ask in Agent mode and Copilot picks the right one, or call it by name with /.

Build a feature

build sets up a CE.SDK Web project and implements the features you ask for. Describe the outcome and Copilot scaffolds and wires it up.

Understand a concept

explain breaks down how CE.SDK works and how the pieces fit together, adapted to your framework and level.

Search the live docs

With the MCP server connected, Copilot answers CE.SDK questions from the current documentation, so guidance reflects the latest release.

Example prompts

Copy a starting point into Copilot Chat. Start a product, extend it feature by feature, or ask how something works before you commit to it.

Scaffold an editor
Build a CE.SDK annotation tool: users mark up screenshots with arrows, boxes, and blur for sensitive data, then export as PNG.
Build a CE.SDK real estate flyer maker with listing templates, photo placeholders, and print-ready PDF export.
Build a CE.SDK video highlight editor: trim clips, add captions, export MP4.
Extend it
Add a template gallery so users start from our designs.
Add AI image generation so users can create visuals from a text prompt.
Add stickers and shapes from our asset library.
Understand it
What export formats does CE.SDK support?
Show me how to configure the CE.SDK toolbar in React.

Tips

Get better results from the first prompt.

  • Prompt with outcomes, not implementation: “users annotate a screenshot and download it” beats “integrate the export API”.
  • MCP tools surface in Copilot’s Agent mode; switch modes if you don’t see them.
  • Skills fire from a plain-language request. To call one directly, type / and pick it by name, such as /build.
  • Commit .vscode/mcp.json so your whole team shares the connection.
  • Run skills and MCP together: skills for scaffolding, the MCP server for live docs search.
VS Code Copilot

Ship a working editor

Install the skills, paste a prompt, and Copilot scaffolds a CE.SDK project you can run.

VS Code Copilot FAQs

No. Switch Copilot Chat to Agent mode, describe the product you want in plain language, and Copilot scaffolds the project and tells you how to run it. You review the result in the browser and keep prompting.

You need a GitHub Copilot subscription for Copilot itself. The CE.SDK skills and MCP server are free and need no API key, and the project Copilot scaffolds runs locally without a license key. Licensing only comes in when you deploy what you built.

Both, ideally. Agent Skills bundle the docs and starter kits offline so Copilot can scaffold autonomously. The MCP server adds real-time documentation search on top, so answers stay current between releases.

Make sure Copilot Chat is in Agent mode and reload the window; MCP tools only surface in Agent mode.