GitHub Copilot (VS Code)

Set up GitHub Copilot in VS Code or the terminal to use Avo's documentation.

Quick use

In chat window type this and VSCode will use Avo's docs map to generate code.

#fetch https://docs.avohq.io/2.0/docs-map.md

Project-level permanent setup

Copilot automatically applies the .github/copilot-instructions.md file to every chat request. Add a line pointing to Avo's docs:

When working with Avo, use the docs at https://docs.avohq.io/2.0/docs-map.md as a reference.

Or save the docs in your repo and reference the file from .github/copilot-instructions.md:

curl -L https://docs.avohq.io/2.0/docs-map.md --create-dirs -o docs/avo-docs-map.md

VSCode also reads the AGENTS.md file from your workspace root, if you prefer the cross-tool standard.

MCP server

MCP (Model Context Protocol) connects coding agents to external tools and context. Copilot can use MCP servers to look up documentation while working.

I suggest using Context7 MCP server which provides many libraries including Avo's docs.

Create a .vscode/mcp.json file in your project (or run the MCP: Add Server command from the Command Palette) with this content:

// .vscode/mcp.json
{
  "servers": {
    "Context7": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp@latest"]
    }
  }
}

Now in Agent Mode you can ask AI anything about Avo, and write use context7 at the end of your prompt.

For example:

create a new Avo resource for a product model. use context7

GitHub Copilot CLI

Copilot CLI also reads .github/copilot-instructions.md and AGENTS.md, so the project instructions above work in the terminal too.

Add Context7 to the CLI separately; .vscode/mcp.json configures the editor, not the CLI:

copilot mcp add context7 -- npx -y @upstash/context7-mcp

Use /mcp in an interactive session to check that the server is available, then include use context7 in your Avo prompt. See GitHub's Copilot CLI MCP setup for configuration scopes.