Limited ecosystem outside AWS
Cold starts
GraphQL feature gaps
Imagine you are building an e-commerce API. Your AppSync repository would organize assets like this: appsync repo
Every push to feature/order-validation deploys a new staging environment, runs the full query suite, and merges only if all checks pass.
A pipeline resolver chains multiple functions together. In your repo, store each pipeline function in functions/ and the main resolver in resolvers/pipelines/. Example use case: fetch user profile → enrich with permissions → fetch posts.
Verdict: Start with a monorepo inside a dedicated appsync-repo. If you outgrow it, split Lambda resolvers into separate repos but keep the schema/ and resolvers/ centralized. Limited ecosystem outside AWS
When searching for "AppSync repo," do not miss the official examples. AWS maintains several public GitHub repos:
Clone these repos, study their structure, and adapt them to your needs.
Specify your license (MIT, Apache-2.0, etc.) Cold starts
If you want, I can:
Which of those would you like next?
One pain point of AppSync is local testing. Your repo should contain tooling to mitigate this.