メニュー
Manage database schema changes with Alembic. Use when: creating migrations, adding tables or columns, checking migration status, rolling back, or seeding the database.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Clerk authentication router. Use when user asks about adding authentication, setting up Clerk, custom sign-in flows, Swift or native iOS auth, native Android auth, Next.js patterns, organizations, syncing users, or testing. Automatically routes to the specific skill based on their task.
Understand and customize authentication in this project. Use when: adding auth to routes, swapping auth providers (Clerk, Auth0, custom SSO), debugging auth issues, or understanding the UserInfo contract and get_current_user dependency.
Build, push, and deploy to AWS using Docker, ECR, and Terraform. Use when: deploying to dev/staging/prod, checking deploy status, rolling back, or running the deploy checklist.
Monitor and auto-fix the development environment. Use when: debugging backend/frontend errors, watching log streams, diagnosing startup failures, or running make dev.
Discover, evaluate, and create project-specific skills. Use when: the user needs a capability not covered by built-in skills, wants to find patterns for a specific tool or service, or asks about extending the skills directory. Also use when asking 'how do I add a new skill.'
Generate project-specific Agent Skills for any codebase by interviewing the developer about their conventions. Use when: the user wants to add skills to an existing project, asks 'how do I make my agents understand my codebase,' wants to create a .skills/ directory, or says 'generate skills for my project.'
| name | db-migrate |
| description | Manage database schema changes with Alembic. Use when: creating migrations, adding tables or columns, checking migration status, rolling back, or seeding the database. |
Manage database schema changes with Alembic in SerpentStack.
docker compose up -d postgrescd backend && uv syncAfter modifying or adding a SQLAlchemy model:
cd backend && uv run alembic revision --autogenerate -m "describe the change"
The migration description should be concise and specific, e.g., "add projects table", "add email column to users", "create index on tasks.status".
Review the generated file in backend/migrations/versions/. Verify:
upgrade() function contains the expected op.create_table, op.add_column, or op.create_index calls.downgrade() function is the inverse of upgrade.Apply all pending migrations:
cd backend && uv run alembic upgrade head
Apply only the next migration:
cd backend && uv run alembic upgrade +1
See the current revision:
cd backend && uv run alembic current
See full migration history:
cd backend && uv run alembic history --verbose
Roll back the last migration:
cd backend && uv run alembic downgrade -1
Roll back to a specific revision:
cd backend && uv run alembic downgrade <revision_id>
Full workflow:
backend/app/models/{name}.py — inherit from Base (provides id, created_at, updated_at).backend/app/models/__init__.py so Alembic detects it.cd backend && uv run alembic revision --autogenerate -m "add {name}s table".cd backend && uv run alembic upgrade head.backend/app/models/{name}.py -- add the new column.cd backend && uv run alembic revision --autogenerate -m "add {column} to {name}s".op.add_column is present.cd backend && uv run alembic upgrade head.| Problem | Solution |
|---|---|
Target database is not up to date | Run uv run alembic upgrade head first |
Can't locate revision | Check alembic.ini for correct script_location |
| Phantom diffs in autogenerate | Compare model against DB schema manually; add to Alembic's exclude list if needed |
relation already exists | The migration was partially applied. Check alembic_version table and fix manually |
| Migration conflicts (multiple heads) | Run uv run alembic merge heads -m "merge migrations" |
ModuleNotFoundError on model import | Ensure model is imported in backend/app/models/__init__.py |
| Docker not running | Testcontainers and local Postgres both require Docker Desktop to be running |
Populate the database with sample development data:
make seed
This runs the async seed CLI command at backend/app/cli/seed.py. The seed script is idempotent — running it multiple times will not create duplicates (it checks for existing rows first).