FlybyOps

Connect an AI assistant

Give an AI assistant (like ChatGPT or Claude) secure access to your workspace so you can ask questions in plain English — “which flights last month had incidents?”, “list the open risks on the Riverside project” — and, if you allow it, take actions like filing an incident.

~2 minutes. Read-only by default — you decide what it can change. The assistant only ever sees what you can see.

Step 1

Connect ChatGPT or Claude (one-click)

The easiest way — no API key. Add FlybyOps as a connector in your AI app; you'll be sent to FlybyOps to sign in, pick a workspace, and approve.

Beta:this is the manual way to add FlybyOps, using ChatGPT's developer mode. We're publishing FlybyOps to the ChatGPT app marketplace — once it's live you'll be able to install it in one click, no developer mode required.

Requires a paid ChatGPT plan (Plus, Pro, Team, or Enterprise).

In ChatGPT, open Settings → Apps & Connectors → Advanced settings and turn on Developer mode.
Back on Apps & Connectors, choose Create(the “create an app” flow) and paste the server URL below. Leave any OAuth Client ID / Secret fields blank — ChatGPT registers itself automatically.
Create it, then Connect— you'll be sent to FlybyOps to sign in, pick a workspace, and approve.

MCP server URL:: https://ai.flybyops.com/mcp

If your version of ChatGPT asks for an OAuth Client ID, use 8fefe280-7062-4757-8330-469ef8ded2a5and leave the secret blank — it's a shared, non-secret identifier, not needed for the automatic flow.

After you connect, you'll be sent to FlybyOps to sign in, pick the workspace, and approve — read-only unless you tick write permissions on that screen. Manage or disconnect later under Settings → API Keys → Connected apps.

The FlybyOps consent screen: pick a workspace and approve the connector

Step 2

Or connect a CLI / your own code (API key)

For Gemini CLI, scripts, or the OpenAI API, use a personal API key instead. In FlybyOps go to Settings → API Keys → New API key, tick any write permissions you want (leave all unchecked for read-only), then copy the key once— it's shown only once. Then add the server:

Add FlybyOps to Gemini CLI in ~/.gemini/settings.json (or .gemini/settings.json in your project), then restart the CLI:

{
  "mcpServers": {
    "flybyops": {
      "httpUrl": "https://ai.flybyops.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Replace YOUR_API_KEY with the key you created above.

Any MCP client works — point it at https://ai.flybyops.com/mcp and send your key as a bearer token in the Authorization header.

Step 3

Ask in plain English

Your assistant now has FlybyOps tools (it already knows your platform — projects, jobs, flights, risks, incidents — from a briefing we send on connect). Just ask; it picks the right tools for you. For example:

“Summarise this month's flights.”
“Which open risks are linked to the Riverside project?”
“Show flights that don't have a pilot assigned.”

Chat asking a question and the assistant answering with FlybyOps data

Permissions & safety

Read-only by default: Whether you connect via OAuth or an API key, access is read-only unless you tick write permissions (log a flight, file an incident, update a risk) — on the consent screen for connectors, or when creating the key.
Never more than you: A connection mirrors your role. It can't see or do anything you couldn't do yourself in the dashboard.
Revoke anytime: Settings → API Keys → Connected apps disconnects an OAuth app (ChatGPT/Claude); the API Keys list revokes a key. Either takes effect immediately. Workspace owners can also switch AI access off for everyone from the same screen.

Support

Stuck connecting? Email support@flybyops.com.

Connect an AI assistant

~2 minutes. Read-only by default — you decide what it can change. The assistant only ever sees what you can see.