Compute
Database
Use
db.batch([...]) for atomic multi-step operations. Schema changes belong in server/src/defs/db_schema.ts plus edgespark db generate and edgespark db migrate.
Storage
For large client uploads, use presigned URLs instead of proxying file bodies through EdgeSpark.
Secrets and vars
Deployment
See also
Platform security
How EdgeSpark enforces auth, runtime SQL rules, and storage isolation.
Use storage
Use presigned URLs and bucket declarations to stay within storage limits.