What your agent can do
Read the whole picture. Act where it counts.
Your AI agent works with the same operational data you see in the cockpit — securely, and always within the boundaries you set.
Secure by design
You stay in control of every connection
Connect in a minute
Three steps, no code
Example client config
{
"mcpServers": {
"orbital-os": {
"url": "https://orbitalos.online/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_TOKEN"
}
}
}
}FAQ
MCP, answered
What is the Model Context Protocol (MCP)?
MCP is an open standard that lets AI assistants like Claude, ChatGPT, and Cursor securely connect to external tools and data. Orbital OS ships a built-in MCP server, so any MCP-compatible agent can work directly with your operational data — no custom integration code required.
Which AI agents can connect to Orbital OS?
Any MCP-compatible client. That includes Claude (desktop and web connectors), ChatGPT custom connectors, and Cursor, plus any other tool that speaks the Model Context Protocol. Because it is a standard, you are never locked into one vendor.
Can external agents change my data, or only read it?
Both, and you stay in control. Connected agents can read your maturity, KPIs, cadences, decisions, and issues, and can take a focused set of write actions such as logging a KPI reading or scheduling a cadence. Every action is scoped to a single organization and recorded in your audit log.
How is access secured?
Each organization authenticates with its own secret bearer token. Tokens are scoped to that one organization, can be rotated or disabled at any time, and every read and write is attributed and written to an immutable audit log. MCP access is available on OS Pro, Enterprise, and Consulting plans.
Do I need to write any code to connect?
No. Enable MCP access in your Integrations settings, copy the server URL and token, and paste the provided config into your AI client. The whole setup takes about a minute.