Migration guide completeness check
You're shipping v3 of an SQL-backed data-access library with several schema and API changes. You have an AI draft the migration guide for users upgrading from v2, covering renamed methods, a changed connection-pool config, and a new required index. The draft is thorough-looking. How do you confirm a migration guide is actually complete and that following it end-to-end leaves a user on a working v3 — without trusting the AI's coverage?
audit_migration_guide(removed_symbols: list[str], guide_lines: list[str]) → list[str][["fetchAll","PoolConfig"],["Step 1: run the schema migration to add the v3 columns.","Step 2: replace fetchAll with findMany in every call site.","Step 3: replace PoolConfig with poolOptions in your client setup.","Step 4: create index CONCURRENTLY idx_orders_customer on orders (customer_id).","Step 5: deploy the v3 application.","Rollback: pin back to v2 and drop the new index."]]out[][["fetchAll","PoolConfig"],["Step 1: deploy the v3 package from npm.","Step 2: replace fetchAll with findMany.","Step 3: create index idx_orders_customer on orders (customer_id).","Step 4: run the migration for the new columns."]]out["bad-ordering","missing-replacement:PoolConfig","missing-rollback","unsafe-index"][["fetchAll"],["fetchAll is removed in v3.","Rollback: reinstall v2."]]out["missing-replacement:fetchAll"]Treat the AI’s output as a draft to verify, not an answer to trust. Name the specific flaw and the input that triggers it, say how you’d catch it (tests, edge cases, reading critically), and how you’d re-prompt or decompose to get it right.
Vibe & agentic: describe the solution in plain language (or narrate it) and the coach grades your approach.