by ponywombat on 4/30/2025, 5:51:27 AM
AWS have also made their own cost analysis MCP server https://github.com/awslabs/mcp/tree/main/src/cost-analysis-m...
by mdaniel on 4/30/2025, 1:46:41 AM
How does this work? https://github.com/vantage-sh/vantage-mcp-server?tab=License...
That is extra weird when thinking about the audience who might be Vantage.sh users (and thus have the ability to create the read-only token mentioned elsewhere) but would almost certainly be using it from their workstation, in a commercial context. Sounds like you're trying to keep someone from selling your MCP toy and decided to be cute with the licensing text
by _pdp_ on 4/30/2025, 7:23:54 AM
On a related note, I'm not sure when it became "ok" to leave production credentials scattered across your system in configuration files. So many MCP server examples encourage this pattern, and inevitably, it's going to cause trouble at some point.
by andrenotgiant on 4/25/2025, 3:25:06 PM
What's the difference between connecting an LLM to the data through Vantage vs directly to the AWS cost and usage API's?
by cat-whisperer on 4/25/2025, 3:00:47 PM
This is going to different, as resources end up getting intertwined? or is there a way to standardize it?
by salynchnew on 4/30/2025, 6:12:30 PM
This is dangrous, as I've long held that inscrutible cloud billing was one of the greatest protections we had against a runaway superintelligent AI.
Now we only have poor IAM UX to fall back on.
/s
I know AI is peak hype right now. But it has definitely changed some of our dev workflows already. So we wanted to find a way to let our customers experiment with how they can use AI to make their cloud cost management work more productive.
The MCP Server acts as a connector between LLMs (right now only Claude, Cursor support it but ChatGPT and Google Gemini coming soon) and your cost and usage data on Vantage which supports 20+ cloud infra providers including AWS, Datadog, Mongo, etc. (You have to have a Vantage account to use it since it's using the Vantage API)
Video demo: https://www.youtube.com/watch?v=n0VP2NlUvRU
Repo: https://github.com/vantage-sh/vantage-mcp-server
It's really impressive how capable the latest-gen models are with an MCP server and an API. So far we have found it useful for:
Ad-Hoc questions: "What's our non-prod cloud spend per engineer if we have 25 engineers"
Action plans: "Find unallocated spend and look for clues how it should be tagged"
Multi-tool workflows: "Find recent cost spikes that look like they could have come from eng changes and look for GitHub PR's merged around the same time" (using it in combination with the GitHub MCP)
Thought I'd share, let me know if you have questions.