Oracle MCP Server

Instructions

Implementation Reference

• src/oracle_mcp_server/server.py:811-820 (handler)

  MCP tool call handler branch for 'list_views': extracts optional 'owner' parameter from arguments, calls DatabaseInspector.get_views(owner), formats the views list as JSON, and returns as TextContent.
  elif name == "list_views": owner = arguments.get("owner") views = await self.inspector.get_views(owner) return [ TextContent( type="text", text=json.dumps({"views": views}, indent=2, default=str), ) ]
• src/oracle_mcp_server/server.py:281-317 (helper)

  Core helper method in DatabaseInspector class that executes SQL query against ALL_VIEWS and ALL_TAB_COMMENTS to retrieve views filtered by optional owner, including owner, view_name, and view_comment.
  async def get_views(self, owner: Optional[str] = None) -> List[Dict[str, Any]]: """Get list of views""" conn = await self.connection_manager.get_connection() try: cursor = conn.cursor() query = """ SELECT v.owner, v.view_name, vc.comments as view_comment FROM all_views v LEFT JOIN all_tab_comments vc ON v.owner = vc.owner AND v.view_name = vc.table_name WHERE 1=1 """ params = [] if owner: query += " AND v.owner = :owner" params.append(owner) query += " ORDER BY v.owner, v.view_name" cursor.execute(query, params) views = [] for row in cursor: views.append( {"owner": row[0], "view_name": row[1], "view_comment": row[2]} ) return views finally: conn.close()
• src/oracle_mcp_server/server.py:681-694 (registration)

  Registers the 'list_views' tool with the MCP server in handle_list_tools(), specifying name, description, and input schema with optional 'owner' parameter.
  Tool( name="list_views", description="List all views in the database", inputSchema={ "type": "object", "properties": { "owner": { "type": "string", "description": "Filter by schema owner (optional)", "default": None, } }, }, ),
• src/oracle_mcp_server/server.py:684-693 (schema)

  Pydantic/MCP input schema for 'list_views' tool: object with optional 'owner' string property for filtering views by schema.
  inputSchema={ "type": "object", "properties": { "owner": { "type": "string", "description": "Filter by schema owner (optional)", "default": None, } }, },