Your references.
Your choice of AI.

1. On ChatGPT web, open Plugins → + → Add custom MCP server. Some accounts show this under Apps or developer settings.
2. Name it Things I FW and paste the server URL above.
3. Choose OAuth with automatic/dynamic registration. Leave static client credentials empty.
4. Create and install the plugin. Complete browser sign-in and approve your library or sections.
5. Start a new chat, type @, select things-i-fw, and send the first test prompt.

If a chat says no tools are available, open the plugin’s app entry under Apps 1, then View details. Look for the four read tools. Select Manage → Refresh tools, then open a new chat and select the plugin again.

Official ChatGPT setup / Refresh tools

1. Open Customize → Connectors → Add → Add custom connector. Team accounts may need an owner to add it first.
2. Enter Things I FW and paste the server URL above.
3. Keep OAuth sign-in enabled. If an OAuth client choice appears, select Register automatically. Leave fixed request headers empty.
4. Click Connect, sign in and approve access to your library or selected sections.
5. In a new conversation, open + → Connectors, enable Things I FW and send the first test prompt.

Use this remote connector flow for Desktop and Cowork too. No local desktop configuration file or bridge process is needed. Claude Code has its own instructions below.

Official Claude setup

1. Update VS Code, sign in to GitHub Copilot and open a project with a tool-capable agent session.
2. Open the Command Palette with Cmd+Shift+P on macOS or Ctrl+Shift+P on Windows/Linux. Run MCP: Add Server.
3. Choose HTTP, paste the server URL above, and use things-i-fw as the server ID.
4. For personal use, choose Global → Copilot Global when available. Start the server and complete browser sign-in and library consent when prompted.
5. Enable its tools in Chat: Open Customizations → Tools. Older/Local sessions use Configure Tools in the chat input. Send the first test prompt.

For manual setup, merge this into ~/.copilot/mcp-config.json (or your configured COPILOT_HOME). Keep existing entries.

{
  "mcpServers": {
    "things-i-fw": {
      "type": "http",
      "url": "https://thingsifw.vercel.app/api/mcp"
    }
  }
}

For VS Code’s Local harness, use .vscode/mcp.json with servers as the outer key instead:

{
  "servers": {
    "things-i-fw": {
      "type": "http",
      "url": "https://thingsifw.vercel.app/api/mcp"
    }
  }
}

Copilot cloud agent and code review: GitHub currently documents that these do not support remote MCP OAuth. Use a compatible interactive client for this connector.

Official VS Code setup / Config formats / Copilot support

1. Open Cursor’s Customize / MCP settings (called Tools & MCP in some versions), or open ~/.cursor/mcp.json for personal use.
2. Merge the configuration below into that file. Use .cursor/mcp.json for one project instead.
3. Enable the server and choose Connect / Authenticate when prompted. Finish browser sign-in and library consent.
4. Open an Agent conversation with the server enabled and send the first test prompt.

{
  "mcpServers": {
    "things-i-fw": {
      "url": "https://thingsifw.vercel.app/api/mcp"
    }
  }
}

Official Cursor setup

With Claude Code installed, run this in your terminal. User scope makes the connection available across projects.

claude mcp add --transport http --scope user things-i-fw https://thingsifw.vercel.app/api/mcp

1. Start Claude Code and run /mcp.
2. Select things-i-fw and its authentication option.
3. Finish browser sign-in and library consent. Return to Claude Code, confirm it is connected, and send the first test prompt.

For a project-shared setup, use --scope project. Each teammate still signs in separately. If the name already exists, inspect it with claude mcp get things-i-fw.

Official Claude Code setup

Add the server in your local client’s MCP settings, choose Streamable HTTP, then Authenticate. Or use the terminal:

codex mcp add things-i-fw --url https://thingsifw.vercel.app/api/mcp
codex mcp login things-i-fw

Complete browser sign-in and library consent. If adding the server already completed OAuth, a second login is unnecessary. Start a fresh session, check /mcp where available, and send the first test prompt.

Manual alternative: merge this into ~/.codex/config.toml, then run codex mcp login things-i-fw.

[mcp_servers.things-i-fw]
url = "https://thingsifw.vercel.app/api/mcp"

Local Codex clients on the same host share configuration. ChatGPT web needs its own plugin setup and does not read this file.

Official OpenAI MCP setup

With Gemini CLI installed, merge this into ~/.gemini/settings.json for personal use, or .gemini/settings.json for one project:

{
  "mcpServers": {
    "things-i-fw": {
      "httpUrl": "https://thingsifw.vercel.app/api/mcp"
    }
  }
}

1. Restart Gemini CLI, then run /mcp auth things-i-fw inside its interactive session.
2. Complete browser sign-in and library consent. Use a machine with a browser and a working local callback.
3. Check /mcp for the tools and send the first test prompt.

Use httpUrl; Gemini CLI’s url selects SSE. These instructions apply to Gemini CLI, not the consumer Gemini chat website. Image indexing in Things I FW is a separate feature.

Official Gemini CLI setup