# Connect to this server

Server: `https://app.giljo.ai`  
MCP address: `https://app.giljo.ai/mcp`  
Server name to register: `giljo_hq`

You are an AI coding agent. Follow these steps to connect your own harness to this server.

## Steps

1. Pick the section below that matches the harness you are running in.
2. Run the command(s) in it.
3. Restart or reload the harness if it asks you to, and confirm the server is connected (for example `claude mcp list`, `codex mcp list`, `opencode mcp list`).
4. Call the `giljo_setup` tool. It installs the rest of the integration (skills and agent profiles) for your harness.

## Which sign-in path

This server is reachable over HTTPS or on this machine (localhost), so use **sign-in (OAuth)**. Do not ask the user for an API key.

## Claude Code

```bash
claude mcp add --transport http giljo_hq https://app.giljo.ai/mcp --scope user
```

Then run `/mcp` inside Claude Code and choose to authenticate; a browser window opens for sign-in.

## Codex CLI

```bash
codex mcp add giljo_hq --url https://app.giljo.ai/mcp
```

Codex detects OAuth on add and opens the browser to sign in.

## OpenCode

```bash
opencode mcp add giljo_hq --url https://app.giljo.ai/mcp
opencode mcp auth giljo_hq
```

## Claude Desktop and claude.ai

Add a custom connector in Settings > Connectors with the address `https://app.giljo.ai/mcp`. It signs in through the browser.

## Generic MCP client

Server address (streamable HTTP): `https://app.giljo.ai/mcp`. The server advertises OAuth at `https://app.giljo.ai/.well-known/oauth-authorization-server`; a client that supports MCP OAuth with dynamic client registration needs nothing else.