Testing

Generated bindings should not be treated as correct just because generation completed.

Validation expectations

verify repo-local generation results
run compiler gates where the new bindings are used
rerun downstreams when the changes affect real application graphs
rerun publish preflight when versions or package contents changed

Typical real gate sequence:

regenerate bindings
run relevant repo-local tests
rerun tsonic compiler gates when call surfaces changed
rerun downstream applications when package graphs changed
run wave preflight before publishing

What counts as repo-local testing

Depending on the repo, that can mean:

generation scripts completing cleanly
package-local smoke tests or selftests
local install/overlay checks in downstream workspaces
version-drift checks during publish preflight

Why this is strict

Binding regressions can break:

overload selection
nullable and generic projection
package metadata expectations
downstream builds that only fail after full graph resolution

A green generator run is not enough by itself.