diff --git a/CLAUDE.md b/CLAUDE.md
index 5f9f778..c4e8e07 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -24,7 +24,7 @@ Pre-implementation design docs are archived in `docs/archive/`.
 
 ## Key Rules
 
-- **Database**: Never use raw SQL for schema initialization. Use `drizzle-kit generate` and the migration system. See [docs/database-migrations.md](docs/database-migrations.md).
+- **Database migrations**: Hand-write SQL in `apps/server/drizzle/NNNN_name.sql` AND register it in `apps/server/drizzle/meta/_journal.json` (increment `idx`, set `tag` to filename without `.sql`). Both files must be committed together. Do NOT use `drizzle-kit generate` — snapshots are stale since migration 0008. See [docs/database-migrations.md](docs/database-migrations.md).
 - **Logging**: Use `createModuleLogger()` from `apps/server/logger/index.ts`. Keep `console.log` for CLI user-facing output only.
 - **Hexagonal architecture**: Repository ports in `apps/server/db/repositories/*.ts`, Drizzle adapters in `apps/server/db/repositories/drizzle/*.ts`. All re-exported from `apps/server/db/index.ts`.
 - **tRPC context**: Optional repos accessed via `require*Repository()` helpers in `apps/server/trpc/routers/_helpers.ts`.
diff --git a/docs/database-migrations.md b/docs/database-migrations.md
index 5d149fe..93b0975 100644
--- a/docs/database-migrations.md
+++ b/docs/database-migrations.md
@@ -1,11 +1,11 @@
 # Database Migrations
 
-This project uses [drizzle-kit](https://orm.drizzle.team/kit-docs/overview) for database schema management and migrations.
+This project uses [drizzle-orm](https://orm.drizzle.team/) for database schema management with **hand-written SQL migrations**.
 
 ## Overview
 
 - **Schema definition:** `apps/server/db/schema.ts` (drizzle-orm table definitions)
-- **Migration output:** `apps/server/drizzle/` directory (SQL files + meta journal)
+- **Migration output:** `apps/server/drizzle/` directory (SQL files + `meta/_journal.json`)
 - **Config:** `drizzle.config.ts`
 - **Runtime migrator:** `apps/server/db/ensure-schema.ts` (calls `drizzle-orm/better-sqlite3/migrator`)
 
@@ -13,17 +13,55 @@ This project uses [drizzle-kit](https://orm.drizzle.team/kit-docs/overview) for
 
 On every server startup, `ensureSchema(db)` runs all pending migrations from the `apps/server/drizzle/` folder. Drizzle tracks applied migrations in a `__drizzle_migrations` table so only new migrations are applied. This is safe to call repeatedly.
 
+The migrator discovers migrations via `apps/server/drizzle/meta/_journal.json` — **not** by scanning the filesystem. A migration SQL file that isn't registered in the journal will never be applied.
+
 ## Workflow
 
 ### Making schema changes
 
+**Do NOT use `drizzle-kit generate`** — the snapshots in `meta/` have been stale since migration 0008 and `drizzle-kit generate` will produce incorrect interactive prompts. All migrations since 0008 are hand-written.
+
 1. Edit `apps/server/db/schema.ts` with your table/column changes
-2. Generate a migration:
-   ```bash
-   npx drizzle-kit generate
-   ```
-3. Review the generated SQL in `apps/server/drizzle/NNNN_*.sql`
-4. Commit the migration file along with your schema change
+2. Create a new SQL migration file: `apps/server/drizzle/NNNN_descriptive_name.sql`
+   - Number it sequentially (check the last migration number)
+   - Write the SQL (ALTER TABLE, CREATE INDEX, etc.)
+3. **Register it in the journal**: edit `apps/server/drizzle/meta/_journal.json`
+   - Add a new entry at the end of the `entries` array:
+     ```json
+     {
+       "idx": <next_number>,
+       "version": "6",
+       "when": <unix_timestamp_ms>,
+       "tag": "NNNN_descriptive_name",
+       "breakpoints": true
+     }
+     ```
+   - `idx`: sequential (previous entry's idx + 1)
+   - `tag`: migration filename **without `.sql` extension**
+   - `when`: any timestamp in milliseconds (e.g., previous + 86400000)
+4. Commit **both** the SQL file and the updated `_journal.json` together
+5. Run `npm run build && npm link` to pick up the changes
+
+### Example
+
+Adding a column to an existing table:
+
+```sql
+-- apps/server/drizzle/0032_add_comment_threading.sql
+ALTER TABLE review_comments ADD COLUMN parent_comment_id TEXT REFERENCES review_comments(id) ON DELETE CASCADE;
+CREATE INDEX review_comments_parent_id_idx ON review_comments(parent_comment_id);
+```
+
+```json
+// In meta/_journal.json entries array:
+{
+  "idx": 32,
+  "version": "6",
+  "when": 1772323200000,
+  "tag": "0032_add_comment_threading",
+  "breakpoints": true
+}
+```
 
 ### Applying migrations
 
@@ -31,20 +69,12 @@ Migrations are applied automatically on server startup. No manual step needed.
 
 For tests, the same `ensureSchema()` function is called on in-memory SQLite databases in `apps/server/db/repositories/drizzle/test-helpers.ts`.
 
-### Checking migration status
-
-```bash
-# See what drizzle-kit would generate (dry run)
-npx drizzle-kit generate --dry-run
-
-# Open drizzle studio to inspect the database
-npx drizzle-kit studio
-```
-
 ## Rules
 
-- **Never hand-write migration SQL.** Always use `drizzle-kit generate` from the schema.
+- **Always hand-write migration SQL.** Do not use `drizzle-kit generate` (stale snapshots).
+- **Always register in `_journal.json`.** The migrator reads the journal, not the filesystem.
+- **Commit SQL + journal together.** A migration file without a journal entry is invisible to the migrator.
 - **Never use raw CREATE TABLE statements** for schema initialization. The migration system handles this.
-- **Always commit migration files.** They are the source of truth for database evolution.
 - **Migration files are immutable.** Once committed, never edit them. Make a new migration instead.
-- **Test with `npx vitest run`** after generating migrations to verify they work with in-memory databases.
+- **Test with `npm test`** after creating migrations to verify they work with in-memory databases.
+- **Keep schema.ts in sync.** The schema file is the source of truth for TypeScript types; migrations are the source of truth for database DDL. Both must reflect the same structure.