# Connect Control D to your AI employee Connect to Control D: specialized service integration. Hire an AI employee, connect Control D in one click, and it takes real actions in Control D for you: reading, writing, and following up while you get on with something else. Works over OAuth or any MCP server, and no workflow builder is involved. ## Things people ask their AI employee to do in Control D - Do the repetitive Control D work I keep putting off. - Watch Control D and tell me when something needs my attention. - Pull what I need out of Control D and report back in plain English. ## How connecting Control D works - Hire an AI employee from the marketplace. It arrives knowing the job. - Connect Control D in one click through its standard OAuth screen. No API keys. - Ask for what you want in plain language. It does the work and reports back. ## Why connect Control D to an AI agent? Your AI employee handles the repetitive Control D work, on a schedule or on request, and tells you what it did. Control D is one of thousands of connections, plus any MCP server you already run, so a single request can cross all of them. Category: Other. ## People who connect Control D also connect - [Connect 21risk to your AI agent](/en/integrations/_21risk) - [Connect 2chat to your AI agent](/en/integrations/_2chat) - [Connect API Labz to your AI agent](/en/integrations/api_labz) - [Connect API Ninjas to your AI agent](/en/integrations/api_ninjas) - [Connect API.Bible to your AI agent](/en/integrations/api_bible) - [Connect Ably to your AI agent](/en/integrations/ably) ## Explore - [Every integration](/en/integrations) - [Hire an AI employee](/en/market) - [Pricing](/en/pricing) ## FAQ ### How do I connect Control D to an AI agent? Hire an AI employee, open its Tools tab, find Control D, and click Connect. You authorise access through the standard Control D OAuth screen and the connection is live the moment you approve it. There are no API keys to paste and nothing to deploy. ### Do I have to build a workflow or write code? No. This is not a workflow builder with triggers and branches. You tell your AI employee what you want in plain language, and it decides which Control D actions to run. If you want it to happen every week, say that too. ### What can an AI agent actually do inside Control D? Everything Control D exposes to us: the full action list is on this page, read and write both. Anything Control D itself does not allow is listed as a limit, so you find out here rather than by failing later. ### Is there an MCP server for Control D? You do not need one, the Control D connection is built in. If you already run your own MCP server, your AI employee can use that too, alongside the built-in connection. ### Will it work with my other tools at the same time? Yes, and that is the point. Control D is one of thousands of connections, so a single request can read from one tool, act in Control D, and report back in your chat. ### Is it safe to give an AI agent access to Control D? You connect Control D once at the company level and choose which employees may use it. Sensitive actions can require your approval before they run, and every action the employee takes is logged.