db_schema_get
Retrieve saved database schema snapshots to access table structures and metadata. Specify workspace and connection to get the latest schema or a specific snapshot by ID.
Instructions
Obtiene un esquema guardado. Si snapshot_id es None, devuelve el más reciente
para workspace_id + connection_name.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| workspace_id | Yes | ||
| connection_name | Yes | ||
| snapshot_id | No |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |
Implementation Reference
- schema_engram_mcp/server.py:68-94 (handler)The tool `db_schema_get` is registered and implemented in `server.py`. It handles retrieving a database schema snapshot by ID or the most recent one.
@mcp.tool() def db_schema_get( workspace_id: str, connection_name: str, snapshot_id: int | None = None, ) -> str: """ Obtiene un esquema guardado. Si `snapshot_id` es None, devuelve el más reciente para workspace_id + connection_name. """ conn = _get_conn() if snapshot_id is not None: row = storage.get_snapshot_by_id(conn, snapshot_id) if row is None: return json.dumps({"error": "snapshot not found"}, ensure_ascii=False) if row["workspace_id"] != workspace_id.strip() or row["connection_name"] != connection_name.strip(): return json.dumps( {"error": "snapshot_id no coincide con workspace/connection"}, ensure_ascii=False, ) else: row = storage.get_latest( conn, workspace_id=workspace_id.strip(), connection_name=connection_name.strip() ) if row is None: return json.dumps({"error": "no hay instantáneas"}, ensure_ascii=False) return json.dumps(row, ensure_ascii=False)