request.json

{
    "name": "sample-data-processing-workflow",
    "description": "A sample workflow demonstrating data processing with approval steps",
    "yaml_definition": {
      "schemaVersion": "1.0.0",
      "version": 1,
      "metadata": {
        "name": "sample-data-processing-workflow",
        "description": "A sample workflow demonstrating data processing with approval steps",
        "tags": [
          "data-processing",
          "sample",
          "approval"
        ],
        "owner": "data-team",
        "timeout": "PT2H"
      },
      "triggers": [
        {
          "type": "manual"
        }
      ],
      "inputs": {
        "dataSource": {
          "type": "string",
          "description": "Source URL for data to process",
          "required": true
        },
        "batchSize": {
          "type": "integer",
          "description": "Number of records to process in each batch",
          "default": 100,
          "minimum": 1,
          "maximum": 1000
        },
        "approverEmail": {
          "type": "string",
          "description": "Email of the person who should approve processing",
          "required": true
        }
      },
      "variables": {
        "processingThreshold": 500,
        "maxRetries": 3
      },
      "secrets": {
        "api_token": {
          "secretId": "data-api-token",
          "type": "bearer_token"
        }
      },
      "workflow": {
        "activities": [
          {
            "id": "fetch_data",
            "name": "Fetch Data from API",
            "type": "task",
            "task": {
              "executor": "api",
              "config": {
                "method": "GET",
                "url": "https://api.example.com/data",
                "headers": {
                  "Content-Type": "application/json",
                  "Authorization": "Bearer ${secrets.api_token}"
                },
                "queryParams": {
                  "source": "${input.dataSource}",
                  "limit": "${input.batchSize}"
                }
              },
              "outputs": {
                "records": "$.data.records",
                "totalCount": "$.data.total_count"
              }
            },
            "timeout": "PT10M",
            "retryPolicy": {
              "maxAttempts": 3,
              "backoff": "exponential",
              "initialInterval": "PT5S",
              "retryableErrors": ["NETWORK_ERROR", "TIMEOUT", "500"]
            }
          },
          {
            "id": "approval_check",
            "name": "Approval for Large Dataset Processing",
            "type": "task",
            "condition": "${fetch_data.output.totalCount > variables.processingThreshold}",
            "requiresApproval": true,
            "approval": {
              "approvers": ["${input.approverEmail}"],
              "prompt": "Dataset contains ${fetch_data.output.totalCount} records. Approve processing?",
              "timeout": "PT1H",
              "onTimeout": "reject",
              "metadata": {
                "recordCount": "${fetch_data.output.totalCount}",
                "dataSource": "${input.dataSource}",
                "estimatedProcessingTime": "PT30M"
              }
            },
            "task": {
              "executor": "script",
              "config": {
                "language": "python",
                "code": "print('Approval granted for processing')"
              }
            }
          },
          {
            "id": "process_data",
            "name": "Process Data with AI Agent",
            "type": "task",
            "task": {
              "executor": "agentic",
              "config": {
                "agent": "mcp://data-processing-agent",
                "tools": ["data_validator", "data_transformer", "quality_checker"],
                "model": "claude-3-sonnet",
                "prompt": "Process the provided data records: validate format, transform to standard schema, and check quality. Return 
  processed results with quality metrics."
              },
              "inputs": {
                "rawData": "${fetch_data.output.records}",
                "batchSize": "${input.batchSize}"
              },
              "outputs": {
                "processedRecords": "$.processed_data",
                "qualityScore": "$.quality_metrics.score",
                "errors": "$.errors"
              }
            },
            "timeout": "PT30M",
            "retryPolicy": {
              "maxAttempts": 2,
              "backoff": "fixed",
              "initialInterval": "PT1M"
            }
          },
          {
            "id": "parallel_outputs",
            "name": "Generate Outputs in Parallel",
            "type": "parallel",
            "branches": [
              {
                "id": "save_to_database",
                "name": "Save to Database",
                "type": "task",
                "task": {
                  "executor": "connector",
                  "config": {
                    "connectorId": "postgres-db",
                    "operation": "bulk_insert",
                    "parameters": {
                      "table": "processed_data",
                      "records": "${process_data.output.processedRecords}"
                    }
                  },
                  "outputs": {
                    "insertedCount": "$.rows_affected"
                  }
                }
              },
              {
                "id": "generate_report",
                "name": "Generate Summary Report",
                "type": "task",
                "task": {
                  "executor": "script",
                  "config": {
                    "language": "python",
                    "code": "import json\nfrom datetime import datetime\n\ndata = json.loads(input_data)\nreport = {\n  'timestamp': 
  datetime.now().isoformat(),\n  'records_processed': len(data['records']),\n  'quality_score': data['quality_score'],\n  'source': 
  data['source']\n}\nprint(json.dumps(report))"
                  },
                  "inputs": {
                    "input_data": {
                      "records": "${process_data.output.processedRecords}",
                      "quality_score": "${process_data.output.qualityScore}",
                      "source": "${input.dataSource}"
                    }
                  },
                  "outputs": {
                    "report": "$"
                  }
                }
              }
            ]
          },
          {
            "id": "join_results",
            "name": "Wait for All Outputs",
            "type": "join",
            "join": {
              "branches": ["save_to_database", "generate_report"],
              "strategy": "all",
              "timeout": "PT10M",
              "aggregateOutputs": true
            }
          }
        ]
      }
    },
    "is_enabled": true
  }