github copilot.md
Install Lightsage MCP in GitHub Copilot
Connect GitHub Copilot to Lightsage MCP from VS Code or Copilot CLI.
Use this guide for GitHub Copilot in VS Code and GitHub Copilot CLI. Both can connect to remote HTTP MCP servers.
1. Get your Lightsage API key
Open the Lightsage dashboard, copy your API key, and store it as LIGHTSAGE_API_KEY:
export LIGHTSAGE_API_KEY="ls_cli_your_api_key"
Do not put your API key in shared project config, screenshots, issue reports, or public repositories.
2. Add Lightsage MCP to GitHub Copilot in VS Code
Open the command palette and run MCP: Open User Configuration. Add Lightsage to your user mcp.json:
{
"inputs": [
{
"id": "lightsage-api-key",
"type": "promptString",
"description": "Lightsage API key",
"password": true
}
],
"servers": {
"lightsage": {
"type": "http",
"url": "https://mcp.lightsage.com/mcp",
"headers": {
"X-Lightsage-Api-Key": "${input:lightsage-api-key}"
}
}
}
}
Keep this in user configuration unless your team has a shared secret-management pattern for workspace MCP config.
3. Optional: add Lightsage MCP to Copilot CLI
For Copilot CLI, add Lightsage with the MCP command:
copilot mcp add --transport http \
--header "X-Lightsage-Api-Key: ${LIGHTSAGE_API_KEY}" \
lightsage https://mcp.lightsage.com/mcp
You can also add Lightsage to ~/.copilot/mcp-config.json:
{
"mcpServers": {
"lightsage": {
"type": "http",
"url": "https://mcp.lightsage.com/mcp",
"headers": {
"X-Lightsage-Api-Key": "ls_cli_your_api_key"
},
"tools": ["*"]
}
}
}
Use this only for private user config. Do not commit static API keys.
4. Verify GitHub Copilot tools
In VS Code, run MCP: List Servers and start the lightsage server if it is not already running.
In Copilot CLI, run:
copilot mcp list
Then run a read-only prompt:
Use the Lightsage MCP server to check my Lightsage auth status.
See the VS Code MCP configuration reference and GitHub Copilot CLI MCP docs.