In short: we take your OpenAPI spec file and generate idiomatic, best in class SDKs in various languages, a highly customizable docs product (for your API and SDKs, with neat specific examples ready to be copy pasted), MCP servers, CLI clients, terraform providers.
Going further into it: the expected user experience for your team is that you create a PR in your own API repo, a GitHub action triggers builds for everything and gives you a summary via PR comments where you can directly see diagnostic feedback, see the exact diff for each SDKs, provide the commit message for your end users. Once your PR is merged we push changes to all your SDK/docs repos and prepare a release PR ready for your team to review and merge. You merge it, everything gets released to your end users.
Now what we build goes way further than that: we have a web platform where you can live edit your Stainless config file and preview your SDKs, a fairly complex diagnostic system, a really cool system that allows you to add your own custom code on top of any generated SDK directly via git — the whole repo is something you can modify to your wishes, we keep track of your custom changes and always reapply on top of the latest codegen output. And a lot of other features (I’m biased because I designed and implemented the public version but I personally really like our spec transforms, they let you apply changes to your spec file downstream, just by modifying your stainless config file).
Does that make sense?
This is very cool, congratulations!
Thank you!
Do you work at stainless?
clicking on their profile will answer your question.