What is doit-mcp-server?

DoiT MCP Server lets AI assistants like Claude tap directly into your DoiT cloud management platform. Instead of manually digging through dashboards to check cloud spending, pull up invoices, or review support tickets, you can just ask your AI assistant in plain language and it fetches the data for you. Under the hood, it uses something called the Model Context Protocol, which is essentially a standard way for AI chatbots to securely connect to external data sources. You install a small configuration that points the server to your DoiT account using an API key. Once connected, the AI gains access to dozens of tools: it can list your cloud assets, run cost analysis reports, check budgets and alerts, look up anomalies in your cloud resources, retrieve invoices, and even create or update support tickets. You can also ask it to query DoiT's built-in cloud cost expert, called AVA, for optimization advice. This is built for teams already using DoiT to manage their cloud infrastructure across providers like Google Cloud and AWS. A founder trying to understand why this month's cloud bill spiked, a PM checking whether a support ticket has been updated, or a finance lead pulling up historical invoices could all just ask Claude rather than navigating the DoiT console themselves. It turns a multi-step dashboard visit into a single conversational question. The server can run either locally on your machine or connect to a hosted version, and it works with several AI clients including Claude Desktop, Cursor, and AWS Q CLI. The project is written in TypeScript and also exposes a core package that other developers can use to build custom integrations if they want to handle the connection layer themselves.