Schema / Type Migration
Plan and execute a multi-file schema or type migration safely — changing a data model, type definition, or interface across the codebase while keeping everything compiling and consistent. Plans the migration FIRST — the before/after shape, every call-site and file touched, the ordering that never leaves the tree broken, and the backout — then applies the change file-by-file, then a reviewer verifies the tree still builds and nothing referencing the old shape remains. Covers find-all-references, safe edit ordering, backward-compatibility shims, codemod-style changes, and compile/test-green verification.
How it runs
| # | Step | Who runs it | What happens |
|---|---|---|---|
| 1 | Plan the migration | Planner | lock before/after, touched files, ordering, backout |
| 2 | Apply the migration | Developer | edit each touched file in the planned order |
| 3 | Verify the tree | Reviewer | confirm it compiles, tests pass, old shape is gone |
| 4 | Evaluate | Reviewer | Grade the deliverable against every acceptance criterion. All pass → finish; any fail → loop back and fix the gap. |
| 5 | Finish | Developer | All acceptance criteria met. Stamp a short summary and report DONE. |
Say something like "migrate a schema" or "change a type everywhere" or "multi-file migration" or "rename a model across the codebase" or "type migration" in chat to start it.