feat: add Alembic database migrations

README.md

@@ -29,10 +29,19 @@ This setup involves the setting of both the frontend and backend components. We
 
 #### Database Schema
 
-The database automatically creates the following tables:
-- `conversations` - Stores conversation metadata (id, user_id, title, timestamps)
+The database uses [Alembic](https://alembic.sqlalchemy.org/) for schema migrations. On startup, the backend automatically applies any pending migrations. The following tables are managed:
+- `conversations` - Stores conversation metadata (uuid, title, timestamps)
 - `messages` - Stores individual messages within conversations
 
+**Database Migration Commands** (from `backend/`):
+```bash
+make db-migrate                     # Apply pending migrations
+make db-revision msg="description"  # Generate a new migration from model changes
+make db-downgrade                   # Roll back one migration
+make db-history                     # Show migration history
+make db-current                     # Show current revision
+```
+
 #### Setting Up PostgreSQL Database Variables
 
 The backend uses PostgreSQL for conversation persistence. Configure these database variables in your `.env` file:

backend/Makefile

@@ -33,3 +33,23 @@ mcp:
 .PHONY: chat
 chat:
 	@uv run python -m chatbot
+
+.PHONY: db-migrate
+db-migrate:
+	@uv run alembic upgrade head
+
+.PHONY: db-revision
+db-revision:
+	@uv run alembic revision --autogenerate -m "$(msg)"
+
+.PHONY: db-downgrade
+db-downgrade:
+	@uv run alembic downgrade -1
+
+.PHONY: db-history
+db-history:
+	@uv run alembic history
+
+.PHONY: db-current
+db-current:
+	@uv run alembic current

backend/alembic.ini

@@ -0,0 +1,43 @@
+[alembic]
+script_location = src/database/alembic
+prepend_sys_path = .
+version_path_separator = os
+
+# sqlalchemy.url is set programmatically in env.py
+sqlalchemy.url =
+
+[post_write_hooks]
+
+[loggers]
+keys = root,sqlalchemy,alembic
+
+[handlers]
+keys = console
+
+[formatters]
+keys = generic
+
+[logger_root]
+level = WARN
+handlers = console
+qualname =
+
+[logger_sqlalchemy]
+level = WARN
+handlers =
+qualname = sqlalchemy.engine
+
+[logger_alembic]
+level = INFO
+handlers =
+qualname = alembic
+
+[handler_console]
+class = StreamHandler
+args = (sys.stderr,)
+level = NOTSET
+formatter = generic
+
+[formatter_generic]
+format = %(levelname)-5.5s [%(name)s] %(message)s
+datefmt = %H:%M:%S

backend/pyproject.toml

@@ -5,6 +5,7 @@ description = "Add your description here"
 readme = "README.md"
 requires-python = ">=3.13"
 dependencies = [
+    "alembic>=1.13.0",
     "asyncpg>=0.30.0",
     "cryptography>=46.0.5",
     "faiss-cpu==1.12.0",

backend/src/database/alembic/env.py

@@ -0,0 +1,52 @@
+from logging.config import fileConfig
+
+from sqlalchemy import engine_from_config, pool
+
+from alembic import context
+
+from src.database.config import get_database_url
+from src.database.models import Base
+
+config = context.config
+
+if config.config_file_name is not None:
+    fileConfig(config.config_file_name, disable_existing_loggers=False)
+
+config.set_main_option("sqlalchemy.url", get_database_url())
+
+target_metadata = Base.metadata
+
+
+def run_migrations_offline() -> None:
+    """Run migrations in 'offline' mode."""
+    url = config.get_main_option("sqlalchemy.url")
+    context.configure(
+        url=url,
+        target_metadata=target_metadata,
+        literal_binds=True,
+        dialect_opts={"paramstyle": "named"},
+    )
+
+    with context.begin_transaction():
+        context.run_migrations()
+
+
+def run_migrations_online() -> None:
+    """Run migrations in 'online' mode."""
+    connectable = engine_from_config(
+        config.get_section(config.config_ini_section, {}),
+        prefix="sqlalchemy.",
+        poolclass=pool.NullPool,
+    )
+
+    with connectable.connect() as connection:
+        context.configure(connection=connection, target_metadata=target_metadata)
+
+        with context.begin_transaction():
+            context.run_migrations()
+
+
+if context.is_offline_mode():
+    run_migrations_offline()
+else:
+    run_migrations_online()

backend/src/database/alembic/script.py.mako

@@ -0,0 +1,26 @@
+"""${message}
+
+Revision ID: ${up_revision}
+Revises: ${down_revision | comma,n}
+Create Date: ${create_date}
+
+"""
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+${imports if imports else ""}
+
+# revision identifiers, used by Alembic.
+revision: str = ${repr(up_revision)}
+down_revision: Union[str, None] = ${repr(down_revision)}
+branch_labels: Union[str, Sequence[str], None] = ${repr(branch_labels)}
+depends_on: Union[str, Sequence[str], None] = ${repr(depends_on)}
+
+
+def upgrade() -> None:
+    ${upgrades if upgrades else "pass"}
+
+
+def downgrade() -> None:
+    ${downgrades if downgrades else "pass"}

backend/src/database/alembic/versions/0001_initial_schema.py

@@ -0,0 +1,56 @@
+"""Initial schema
+
+Revision ID: 0001
+Revises:
+Create Date: 2026-03-01
+
+"""
+
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision: str = "0001"
+down_revision: Union[str, None] = None
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "conversations",
+        sa.Column("uuid", sa.Uuid(), primary_key=True),
+        sa.Column("title", sa.String(500), nullable=True),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
+        sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
+    )
+
+    op.create_table(
+        "messages",
+        sa.Column("uuid", sa.Uuid(), primary_key=True),
+        sa.Column(
+            "conversation_uuid",
+            sa.Uuid(),
+            sa.ForeignKey("conversations.uuid", ondelete="CASCADE"),
+            nullable=False,
+        ),
+        sa.Column("role", sa.String(50), nullable=False),
+        sa.Column("content", sa.Text(), nullable=False),
+        sa.Column("context_sources", sa.JSON(), nullable=True),
+        sa.Column("tools", sa.JSON(), nullable=True),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
+    )
+
+    op.create_index(
+        "ix_messages_conversation_uuid",
+        "messages",
+        ["conversation_uuid"],
+    )
+
+
+def downgrade() -> None:
+    op.drop_index("ix_messages_conversation_uuid", table_name="messages")
+    op.drop_table("messages")
+    op.drop_table("conversations")

backend/src/database/config.py

@@ -2,6 +2,7 @@
 
 import os
 import logging
+from pathlib import Path
 from typing import Generator, Optional
 from sqlalchemy import create_engine, inspect, text, Engine
 from sqlalchemy.orm import sessionmaker, Session
@@ -40,6 +41,32 @@ def is_database_available() -> bool:
         return False
 
 
+def run_migrations() -> None:
+    """Run Alembic migrations to bring the database schema up to date.
+
+    Auto-detects pre-Alembic databases (app tables exist but no alembic_version
+    table) and stamps them with the current head revision so that future
+    migrations apply cleanly.
+    """
+    from alembic.config import Config
+    from alembic import command
+
+    alembic_ini = Path(__file__).resolve().parents[2] / "alembic.ini"
+    alembic_cfg = Config(str(alembic_ini))
+
+    # Stamp pre-Alembic databases so migrations don't try to recreate tables
+    inspector = inspect(engine)
+    existing_tables = inspector.get_table_names()
+    has_app_tables = "conversations" in existing_tables
+    has_alembic = "alembic_version" in existing_tables
+
+    if has_app_tables and not has_alembic:
+        logger.info("Pre-Alembic database detected — stamping with current head.")
+        command.stamp(alembic_cfg, "head")
+    else:
+        command.upgrade(alembic_cfg, "head")
+
+
 def init_database() -> bool:
     global engine, SessionLocal, _db_initialized
 
@@ -64,15 +91,14 @@ def init_database() -> bool:
             logger.warning("Database is not available. Will retry on next access.")
             return False
 
-        inspector = inspect(engine)
-        existing_tables = inspector.get_table_names()
-
-        if not existing_tables or "conversations" not in existing_tables:
-            logger.info("Initializing database tables...")
+        try:
+            run_migrations()
+        except Exception:
+            logger.warning(
+                "Alembic migration failed; falling back to create_all for table creation.",
+                exc_info=True,
+            )
             Base.metadata.create_all(bind=engine)
-            logger.info("Database tables created successfully")
-        else:
-            logger.debug("Database tables already exist")
 
         _db_initialized = True
         return True