We’ve built Spacelift Intent to make it much easier to build ad-hoc cloud infrastructure with AI. It’s an MCP server that uses Terraform/OpenTofu providers under the hood to talk directly to your cloud provider, and lets your AI agent create and modify cloud resources.
You can either use the open-source version which is just a binary, or the Spacelift-hosted version[0] as a remote MCP server (there you also get stuff like policies, audit history, and credential management).
Compared to clickops/raw cloud cli invocations it also keeps track of all managed resources. This is especially useful across e.g. Claude Code sessions, as even though the conversation context is gone, the assistant can easily read the current state of managed resources, and you can pick up where you left off. This also makes it easy to later dump it all into a tf config + statefile.
Hope you will give it a try, and curious to hear your thoughts!
cube2222•4mo ago
We’ve built Spacelift Intent to make it much easier to build ad-hoc cloud infrastructure with AI. It’s an MCP server that uses Terraform/OpenTofu providers under the hood to talk directly to your cloud provider, and lets your AI agent create and modify cloud resources.
You can either use the open-source version which is just a binary, or the Spacelift-hosted version[0] as a remote MCP server (there you also get stuff like policies, audit history, and credential management).
Compared to clickops/raw cloud cli invocations it also keeps track of all managed resources. This is especially useful across e.g. Claude Code sessions, as even though the conversation context is gone, the assistant can easily read the current state of managed resources, and you can pick up where you left off. This also makes it easy to later dump it all into a tf config + statefile.
Hope you will give it a try, and curious to hear your thoughts!
[0]: https://spacelift.io/intent