GitHub Copilot

Works with the CLI alone. No gander.md account.

GitHub Copilot is GitHub's AI pair programmer. It works in VS Code, JetBrains, the Copilot CLI, and cloud agents, suggesting and applying code next to the files you already have.

GitHub Copilot in VS Code, Copilot CLI, and Copilot cloud agent load Agent Skills from .github/skills, .agents/skills, .claude/skills, and the user dests ~/.copilot/skills, ~/.agents/skills, and ~/.claude/skills. gander skill already links the last two. Type /skills in Copilot Chat, or Chat: Open Customizations, to see Gander.

1. Install the CLI

curl -fsSL https://release.gander.md/install.sh | bash
gander --version

After Homebrew or a source build, run gander skill.

2. Confirm the skill landed

ls -l ~/.agents/skills/gander/SKILL.md
ls -l ~/.claude/skills/gander/SKILL.md

Optional Copilot-native dest:

mkdir -p ~/.copilot/skills
ln -sfn ~/.gander/skill ~/.copilot/skills/gander

Reload the VS Code window. In Copilot Chat (agent mode), type /gander or open Configure Skills.

To commit the skill for a repo, copy or symlink it to .github/skills/gander or .agents/skills/gander.

3. Sign up and add MCP

gander signup --email [email protected]

VS Code MCP is .vscode/mcp.json in the workspace, or a user mcp.json. Example:

{
  "servers": {
    "gander": {
      "type": "stdio",
      "command": "gander",
      "args": ["mcp"]
    }
  }
}

Some Copilot builds use mcpServers with command / args instead of servers. Match the file your editor already has. Copilot CLI reads ~/.copilot MCP config; add the same stdio server there if you work in the terminal.

4. Use it

1. In agent mode, ask for a plan as markdown, then gander this file.
2. gander watch plan.md after signup.
3. Send the hosted URL. Reviewers stay in the browser.
4. @agent comments are MCP tools. Copilot edits the local file and replies.

Fallback: gander comments.

5. Verify

gander list
gander comments

In Copilot Chat, confirm /gander and the four Gander MCP tools.

More: Skill, MCP, Hosted watch.