← Retour aux workflows
Workflow n8n

Automatisation n8n : gestion de requêtes API en batch

  • Ce workflow n8n est conçu pour automatiser la gestion des requêtes API en batch, facilitant ainsi le traitement de multiples prompts de manière efficace. Dans un contexte où les entreprises doivent souvent interagir avec des API pour récupérer ou envoyer des données, ce workflow offre une solution robuste pour optimiser ces processus. Par exemple, il peut être utilisé par des équipes de développement ou des data scientists pour soumettre des requêtes à l'API d'Anthropic Claude, vérifier leur statut et récupérer les résultats sans intervention manuelle.
  • Le workflow commence par un déclencheur manuel, permettant à l'utilisateur de lancer le processus à la demande. Ensuite, il soumet un lot de requêtes via le nœud 'Submit batch', suivi par une vérification de l'état des requêtes avec le nœud 'Check batch status'. Les résultats sont ensuite récupérés et analysés grâce à des nœuds tels que 'Get results' et 'Parse response'. Des nœuds conditionnels comme 'If ended processing' permettent de gérer le flux en fonction des résultats obtenus. En intégrant des fonctionnalités de mémoire, comme 'Simple Memory Store', ce workflow assure une gestion efficace des données tout au long du processus.
  • Les bénéfices de cette automatisation n8n sont nombreux : réduction des erreurs humaines, gain de temps considérable et amélioration de la productivité des équipes. En automatisant ces tâches répétitives, les entreprises peuvent se concentrer sur des activités à plus forte valeur ajoutée, tout en garantissant une interaction fluide avec les API.
Tags clés :automatisationAPIn8ngestion de donnéesworkflow
🎯Pour qui ?🧩Problème résolu📋Étapes du flux🛠Guide de personnalisation
Catégorie: Manual · Tags: automatisation, API, n8n, gestion de données, workflow❤ 0

Workflow n8n gestion de données : vue d'ensemble

Schéma des nœuds et connexions de ce workflow n8n, généré à partir du JSON n8n.

React Flow
Besoin d'aide ?

Workflow n8n gestion de données : détail des nœuds

  • Submit batch

    Ce noeud envoie une requête HTTP pour soumettre un lot de données.

  • Check batch status

    Ce noeud vérifie le statut d'un lot en envoyant une requête HTTP.

  • When Executed by Another Workflow

    Ce noeud déclenche l'exécution d'un autre workflow.

  • Get results

    Ce noeud récupère les résultats en envoyant une requête HTTP.

  • Parse response

    Ce noeud traite la réponse reçue en exécutant du code JavaScript.

  • If ended processing

    Ce noeud évalue une condition pour déterminer si le traitement est terminé.

  • Sticky Note

    Ce noeud crée une note autocollante avec un contenu spécifique.

  • Sticky Note1

    Ce noeud crée une autre note autocollante avec un contenu spécifique.

  • Batch Status Poll Interval

    Ce noeud introduit une pause dans le workflow pour un intervalle de temps défini.

  • Sticky Note2

    Ce noeud crée une note autocollante avec une couleur et un contenu spécifiques.

  • Run example

    Ce noeud déclenche manuellement l'exécution du workflow.

  • One query example

    Ce noeud définit des valeurs pour des options spécifiques dans le workflow.

  • Delete original properties

    Ce noeud supprime les propriétés originales d'un ensemble de données.

  • Construct 'requests' array

    Ce noeud construit un tableau de requêtes à partir des données fournies.

  • Set desired 'anthropic-version'

    Ce noeud définit la version souhaitée d'Anthropic pour le traitement.

  • Execute Workflow 'Process Multiple Prompts in Parallel with Anthropic Claude Batch API'

    Ce noeud exécute un workflow spécifique pour traiter plusieurs requêtes en parallèle.

  • Build batch 'request' object for single query

    Ce noeud construit un objet de requête pour une requête unique à l'aide de code JavaScript.

  • Simple Memory Store

    Ce noeud stocke des données en mémoire pour une session donnée.

  • Fill Chat Memory with example data

    Ce noeud remplit la mémoire de chat avec des données d'exemple.

  • Build batch 'request' object from Chat Memory and execution data

    Ce noeud construit un objet de requête à partir de la mémoire de chat et des données d'exécution.

  • Load Chat Memory Data

    Ce noeud charge les données de la mémoire de chat pour utilisation ultérieure.

  • First Prompt Result

    Ce noeud enregistre le résultat du premier prompt dans les données d'exécution.

  • Second Prompt Result

    Ce noeud enregistre le résultat du deuxième prompt dans les données d'exécution.

  • Split Out Parsed Results

    Ce noeud sépare les résultats analysés en fonction d'un champ spécifique.

  • Filter Second Prompt Results

    Ce noeud fusionne deux requêtes d'exemple en un tableau.

  • Append execution data for single query example

    Ce noeud ajoute des données d'exécution pour un exemple de requête unique.

  • Append execution data for chat memory example

    Ce noeud ajoute des données d'exécution pour un exemple de mémoire de chat.

  • Truncate Chat Memory

    Ce noeud tronque la mémoire de chat selon les paramètres spécifiés.

Inscris-toi pour voir l'intégralité du workflow

Inscription gratuite

S'inscrire gratuitementBesoin d'aide ?
{
  "meta": {
    "instanceId": "97d44c78f314fab340d7a5edaf7e2c274a7fbb8a7cd138f53cc742341e706fe7"
  },
  "nodes": [
    {
      "id": "fa4f8fd6-3272-4a93-8547-32d13873bbc1",
      "name": "Submit batch",
      "type": "n8n-nodes-base.httpRequest",
      "position": [
        180,
        40
      ],
      "parameters": {
        "url": "https://api.anthropic.com/v1/messages/batches",
        "method": "POST",
        "options": {},
        "jsonBody": "={ \"requests\": {{ JSON.stringify($json.requests) }} }",
        "sendBody": true,
        "sendQuery": true,
        "sendHeaders": true,
        "specifyBody": "json",
        "authentication": "predefinedCredentialType",
        "queryParameters": {
          "parameters": [
            {}
          ]
        },
        "headerParameters": {
          "parameters": [
            {
              "name": "anthropic-version",
              "value": "={{ $json[\"anthropic-version\"] }}"
            }
          ]
        },
        "nodeCredentialType": "anthropicApi"
      },
      "credentials": {
        "anthropicApi": {
          "id": "ub0zN7IP2V83OeTf",
          "name": "Anthropic account"
        }
      },
      "typeVersion": 4.2
    },
    {
      "id": "2916dc85-829d-491a-a7a8-de79d5356a53",
      "name": "Check batch status",
      "type": "n8n-nodes-base.httpRequest",
      "position": [
        840,
        115
      ],
      "parameters": {
        "url": "=https://api.anthropic.com/v1/messages/batches/{{ $json.id }}",
        "options": {},
        "sendHeaders": true,
        "authentication": "predefinedCredentialType",
        "headerParameters": {
          "parameters": [
            {
              "name": "anthropic-version",
              "value": "={{ $('When Executed by Another Workflow').item.json[\"anthropic-version\"] }}"
            }
          ]
        },
        "nodeCredentialType": "anthropicApi"
      },
      "credentials": {
        "anthropicApi": {
          "id": "ub0zN7IP2V83OeTf",
          "name": "Anthropic account"
        }
      },
      "typeVersion": 4.2
    },
    {
      "id": "1552ec92-2f18-42f6-b67f-b6f131012b3c",
      "name": "When Executed by Another Workflow",
      "type": "n8n-nodes-base.executeWorkflowTrigger",
      "position": [
        -40,
        40
      ],
      "parameters": {
        "workflowInputs": {
          "values": [
            {
              "name": "anthropic-version"
            },
            {
              "name": "requests",
              "type": "array"
            }
          ]
        }
      },
      "typeVersion": 1.1
    },
    {
      "id": "4bd40f02-caf1-419d-8261-a149cd51a534",
      "name": "Get results",
      "type": "n8n-nodes-base.httpRequest",
      "position": [
        620,
        -160
      ],
      "parameters": {
        "url": "={{ $json.results_url }}",
        "options": {},
        "sendHeaders": true,
        "authentication": "predefinedCredentialType",
        "headerParameters": {
          "parameters": [
            {
              "name": "anthropic-version",
              "value": "={{ $('When Executed by Another Workflow').item.json[\"anthropic-version\"] }}"
            }
          ]
        },
        "nodeCredentialType": "anthropicApi"
      },
      "credentials": {
        "anthropicApi": {
          "id": "ub0zN7IP2V83OeTf",
          "name": "Anthropic account"
        }
      },
      "typeVersion": 4.2
    },
    {
      "id": "5df366af-a54d-4594-a1ab-7a9df968101e",
      "name": "Parse response",
      "type": "n8n-nodes-base.code",
      "notes": "JSONL separated by newlines",
      "position": [
        840,
        -160
      ],
      "parameters": {
        "jsCode": "for (const item of $input.all()) {\n  if (item.json && item.json.data) {\n    // Split the string into individual JSON objects\n    const jsonStrings = item.json.data.split('\\n');\n\n    // Parse each JSON string and store them in an array\n    const parsedData = jsonStrings.filter(str => str.trim() !== '').map(str => JSON.parse(str));\n\n    // Replace the original json with the parsed array.\n    item.json.parsed = parsedData;\n  }\n}\n\nreturn $input.all();"
      },
      "notesInFlow": true,
      "typeVersion": 2
    },
    {
      "id": "68aa4ee2-e925-4e30-a7ab-317d8df4d9bc",
      "name": "If ended processing",
      "type": "n8n-nodes-base.if",
      "position": [
        400,
        40
      ],
      "parameters": {
        "options": {},
        "conditions": {
          "options": {
            "version": 2,
            "leftValue": "",
            "caseSensitive": true,
            "typeValidation": "strict"
          },
          "combinator": "and",
          "conditions": [
            {
              "id": "9494c5a3-d093-49c5-837f-99cd700a2f13",
              "operator": {
                "type": "string",
                "operation": "equals"
              },
              "leftValue": "={{ $json.processing_status }}",
              "rightValue": "ended"
            }
          ]
        }
      },
      "typeVersion": 2.2
    },
    {
      "id": "2b974e3b-495b-48af-8080-c7913d7a2ba8",
      "name": "Sticky Note",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -200,
        -720
      ],
      "parameters": {
        "width": 1060,
        "height": 520,
        "content": "### This workflow automates sending batched prompts to Claude using the Anthropic API. It submits multiple prompts at once and retrieves the results.\n\n#### How to use\n\nCall this workflow with array of `requests`\n\n```json\n{\n    \"anthropic-version\": \"2023-06-01\",\n    \"requests\": [\n        {\n            \"custom_id\": \"first-prompt-in-my-batch\",\n            \"params\": {\n                \"max_tokens\": 100,\n                \"messages\": [\n                    {\n                        \"content\": \"Hey Claude, tell me a short fun fact about video games!\",\n                        \"role\": \"user\"\n                    }\n                ],\n                \"model\": \"claude-3-5-haiku-20241022\"\n            }\n        }\n    ]\n}\n```\n"
      },
      "typeVersion": 1
    },
    {
      "id": "928a30b5-5d90-4648-a82e-e4f1a01e47a5",
      "name": "Sticky Note1",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        1200,
        -720
      ],
      "parameters": {
        "width": 980,
        "height": 600,
        "content": "#### Results\n\nThis workflow returns an array of results with custom_ids.\n\n```json\n[\n    {\n        \"custom_id\": \"first-prompt-in-my-batch\",\n        \"result\": {\n            \"message\": {\n                \"content\": [\n                    {\n                        \"text\": \"Did you know that the classic video game Tetris was...\",\n                        \"type\": \"text\"\n                    }\n                ],\n                \"id\": \"msg_01AiLiVZT18XnoBD4r2w9x2t\",\n                \"model\": \"claude-3-5-haiku-20241022\",\n                \"role\": \"assistant\",\n                \"stop_reason\": \"end_turn\",\n                \"stop_sequence\": null,\n                \"type\": \"message\",\n                \"usage\": {\n                    \"cache_creation_input_tokens\": 0,\n                    \"cache_read_input_tokens\": 0,\n                    \"input_tokens\": 45,\n                    \"output_tokens\": 83\n                }\n            },\n            \"type\": \"succeeded\"\n        }\n    }\n]\n```"
      },
      "typeVersion": 1
    },
    {
      "id": "5dcb554e-32df-4883-b5a1-b40305756201",
      "name": "Batch Status Poll Interval",
      "type": "n8n-nodes-base.wait",
      "position": [
        620,
        40
      ],
      "webhookId": "7efafe72-063a-45c6-8775-fcec14e1d263",
      "parameters": {
        "amount": 10
      },
      "typeVersion": 1.1
    },
    {
      "id": "c25cfde5-ab83-4e5a-a66f-8cc9f23a01f6",
      "name": "Sticky Note2",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -160,
        325
      ],
      "parameters": {
        "color": 4,
        "width": 340,
        "height": 620,
        "content": "# Usage example"
      },
      "typeVersion": 1
    },
    {
      "id": "6062ca7c-aa08-4805-9c96-65e5be8a38fd",
      "name": "Run example",
      "type": "n8n-nodes-base.manualTrigger",
      "position": [
        -40,
        625
      ],
      "parameters": {},
      "typeVersion": 1
    },
    {
      "id": "9878729a-123d-4460-a582-691ca8cedf98",
      "name": "One query example",
      "type": "n8n-nodes-base.set",
      "position": [
        634,
        775
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "1ea47ba2-64be-4d69-b3db-3447cde71645",
              "name": "query",
              "type": "string",
              "value": "Hey Claude, tell me a short fun fact about bees!"
            }
          ]
        }
      },
      "typeVersion": 3.4
    },
    {
      "id": "df06c209-8b6a-4b6d-8045-230ebdfcfbad",
      "name": "Delete original properties",
      "type": "n8n-nodes-base.set",
      "position": [
        1528,
        775
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "d238d62b-2e91-4242-b509-8cfc698d2252",
              "name": "custom_id",
              "type": "string",
              "value": "={{ $json.custom_id }}"
            },
            {
              "id": "21e07c09-92e3-41e7-8335-64653722e7e9",
              "name": "params",
              "type": "object",
              "value": "={{ $json.params }}"
            }
          ]
        }
      },
      "typeVersion": 3.4
    },
    {
      "id": "f66d6a89-ee33-4494-9476-46f408976b29",
      "name": "Construct 'requests' array",
      "type": "n8n-nodes-base.aggregate",
      "position": [
        1968,
        625
      ],
      "parameters": {
        "options": {},
        "aggregate": "aggregateAllItemData",
        "destinationFieldName": "requests"
      },
      "typeVersion": 1
    },
    {
      "id": "0f9eb605-d629-4cb7-b9cb-39702d201567",
      "name": "Set desired 'anthropic-version'",
      "type": "n8n-nodes-base.set",
      "notes": "2023-06-01",
      "position": [
        2188,
        625
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "9f9e94a0-304b-487a-8762-d74421ef4cc0",
              "name": "anthropic-version",
              "type": "string",
              "value": "2023-06-01"
            }
          ]
        },
        "includeOtherFields": true
      },
      "notesInFlow": true,
      "typeVersion": 3.4
    },
    {
      "id": "f71f261c-f4ad-4c9f-bd72-42ab386a65e1",
      "name": "Execute Workflow 'Process Multiple Prompts in Parallel with Anthropic Claude Batch API'",
      "type": "n8n-nodes-base.executeWorkflow",
      "notes": "See above",
      "position": [
        2408,
        625
      ],
      "parameters": {
        "options": {
          "waitForSubWorkflow": true
        },
        "workflowId": {
          "__rl": true,
          "mode": "list",
          "value": "xQU4byMGhgFxnTIH",
          "cachedResultName": "Process Multiple Prompts in Parallel with Anthropic Claude Batch API"
        },
        "workflowInputs": {
          "value": {
            "requests": "={{ $json.requests }}",
            "anthropic-version": "={{ $json['anthropic-version'] }}"
          },
          "schema": [
            {
              "id": "anthropic-version",
              "type": "string",
              "display": true,
              "removed": false,
              "required": false,
              "displayName": "anthropic-version",
              "defaultMatch": false,
              "canBeUsedToMatch": true
            },
            {
              "id": "requests",
              "type": "array",
              "display": true,
              "removed": false,
              "required": false,
              "displayName": "requests",
              "defaultMatch": false,
              "canBeUsedToMatch": true
            }
          ],
          "mappingMode": "defineBelow",
          "matchingColumns": [
            "requests"
          ],
          "attemptToConvertTypes": true,
          "convertFieldsToString": true
        }
      },
      "notesInFlow": true,
      "typeVersion": 1.2
    },
    {
      "id": "bd27c1a6-572c-420d-84ab-4d8b7d14311b",
      "name": "Build batch 'request' object for single query",
      "type": "n8n-nodes-base.code",
      "position": [
        1308,
        775
      ],
      "parameters": {
        "jsCode": "// Loop over input items and modify them to match the response example, then return input.all()\nfor (const item of $input.all()) {\n  item.json.params = {\n    max_tokens: item.json.max_tokens,\n    messages: [\n      {\n        content: item.json.query,\n        role: \"user\"\n      }\n    ],\n    model: item.json.model\n  };\n}\n\nreturn $input.all();\n"
      },
      "typeVersion": 2
    },
    {
      "id": "fa342231-ea94-43ab-8808-18c8d04fdaf8",
      "name": "Simple Memory Store",
      "type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
      "position": [
        644,
        595
      ],
      "parameters": {
        "sessionKey": "\"Process Multiple Prompts in Parallel with Anthropic Claude Batch API example\"",
        "sessionIdType": "customKey"
      },
      "typeVersion": 1.3
    },
    {
      "id": "67047fe6-8658-45ba-be61-52cf6115f4e4",
      "name": "Fill Chat Memory with example data",
      "type": "@n8n/n8n-nodes-langchain.memoryManager",
      "position": [
        556,
        375
      ],
      "parameters": {
        "mode": "insert",
        "messages": {
          "messageValues": [
            {
              "message": "You are a helpful AI assistant"
            },
            {
              "type": "user",
              "message": "Hey Claude, tell me a short fun fact about video games!"
            },
            {
              "type": "ai",
              "message": "short fun fact about video games!"
            },
            {
              "type": "user",
              "message": "No, an actual fun fact"
            }
          ]
        }
      },
      "typeVersion": 1.1
    },
    {
      "id": "dbb295b8-01fd-445f-ab66-948442b6c71d",
      "name": "Build batch 'request' object from Chat Memory and execution data",
      "type": "n8n-nodes-base.code",
      "position": [
        1528,
        475
      ],
      "parameters": {
        "jsCode": "const output = [];\n\nfor (const item of $input.all()) {\n  const inputMessages = item.json.messages;\n  const customId = item.json.custom_id;\n  const model = item.json.model;\n  const maxTokens = item.json.max_tokens;\n\n  if (inputMessages && inputMessages.length > 0) {\n    let systemMessageContent = undefined;\n    const transformedMessages = [];\n\n    // Process each message entry in sequence\n    for (const messageObj of inputMessages) {\n      // Extract system message if present\n      if ('system' in messageObj) {\n        systemMessageContent = messageObj.system;\n      }\n      \n      // Process human and AI messages in the order they appear in the object keys\n      // We need to determine what order the keys appear in the original object\n      const keys = Object.keys(messageObj);\n      \n      for (const key of keys) {\n        if (key === 'human') {\n          transformedMessages.push({\n            role: \"user\",\n            content: messageObj.human\n          });\n        } else if (key === 'ai') {\n          transformedMessages.push({\n            role: \"assistant\",\n            content: messageObj.ai\n          });\n        }\n        // Skip 'system' as we already processed it\n      }\n    }\n\n    const params = {\n      model: model,\n      max_tokens: maxTokens,\n      messages: transformedMessages\n    };\n\n    if (systemMessageContent !== undefined) {\n      params.system = systemMessageContent;\n    }\n\n    output.push({\n      custom_id: customId,\n      params: params\n    });\n  }\n}\n\nreturn output;"
      },
      "typeVersion": 2
    },
    {
      "id": "f9edb335-c33d-45fc-8f9b-12d7f37cc23e",
      "name": "Load Chat Memory Data",
      "type": "@n8n/n8n-nodes-langchain.memoryManager",
      "position": [
        932,
        475
      ],
      "parameters": {
        "options": {}
      },
      "typeVersion": 1.1
    },
    {
      "id": "22399660-ebe5-4838-bad3-c542d6d921a3",
      "name": "First Prompt Result",
      "type": "n8n-nodes-base.executionData",
      "position": [
        2848,
        525
      ],
      "parameters": {
        "dataToSave": {
          "values": [
            {
              "key": "assistant_response",
              "value": "={{ $json.result.message.content[0].text }}"
            }
          ]
        }
      },
      "typeVersion": 1
    },
    {
      "id": "0e7f44f4-c931-4e0f-aebc-1b8f0327647f",
      "name": "Second Prompt Result",
      "type": "n8n-nodes-base.executionData",
      "position": [
        2848,
        725
      ],
      "parameters": {
        "dataToSave": {
          "values": [
            {
              "key": "assistant_response",
              "value": "={{ $json.result.message.content[0].text }}"
            }
          ]
        }
      },
      "typeVersion": 1
    },
    {
      "id": "e42b01e0-8fc5-42e1-aa45-aa85477e766b",
      "name": "Split Out Parsed Results",
      "type": "n8n-nodes-base.splitOut",
      "position": [
        1060,
        -160
      ],
      "parameters": {
        "options": {},
        "fieldToSplitOut": "parsed"
      },
      "typeVersion": 1
    },
    {
      "id": "343676b9-f147-4981-b555-8af570374e8c",
      "name": "Filter Second Prompt Results",
      "type": "n8n-nodes-base.filter",
      "position": [
        2628,
        725
      ],
      "parameters": {
        "options": {},
        "conditions": {
          "options": {
            "version": 2,
            "leftValue": "",
            "caseSensitive": true,
            "typeValidation": "strict"
          },
          "combinator": "and",
          "conditions": [
            {
              "id": "9e4b3524-7066-46cc-a365-8d23d08c1bda",
              "operator": {
                "name": "filter.operator.equals",
                "type": "string",
                "operation": "equals"
              },
              "leftValue": "={{ $json.custom_id }}",
              "rightValue": "={{ $('Append execution data for single query example').item.json.custom_id }}"
            }
          ]
        }
      },
      "typeVersion": 2.2
    },
    {
      "id": "c9f5f366-27c4-4401-965b-67c314036fb6",
      "name": "Filter First Prompt Results",
      "type": "n8n-nodes-base.filter",
      "position": [
        2628,
        525
      ],
      "parameters": {
        "options": {},
        "conditions": {
          "options": {
            "version": 2,
            "leftValue": "",
            "caseSensitive": true,
            "typeValidation": "strict"
          },
          "combinator": "and",
          "conditions": [
            {
              "id": "9e4b3524-7066-46cc-a365-8d23d08c1bda",
              "operator": {
                "name": "filter.operator.equals",
                "type": "string",
                "operation": "equals"
              },
              "leftValue": "={{ $json.custom_id }}",
              "rightValue": "={{ $('Append execution data for chat memory example').item.json.custom_id }}"
            }
          ]
        }
      },
      "typeVersion": 2.2
    },
    {
      "id": "0a5b9c3d-665b-4e35-be9e-c8297314969d",
      "name": "Sticky Note3",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        110,
        -100
      ],
      "parameters": {
        "height": 300,
        "content": "## Submit batch request to Anthropic"
      },
      "typeVersion": 1
    },
    {
      "id": "f19813a5-f669-45dd-a446-947a30b02b09",
      "name": "Sticky Note4",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        350,
        -5
      ],
      "parameters": {
        "width": 640,
        "height": 300,
        "content": "## Loop until processing status is 'ended'"
      },
      "typeVersion": 1
    },
    {
      "id": "9f424fce-5610-4b85-9be6-4c2c403002db",
      "name": "Sticky Note5",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        500,
        -200
      ],
      "parameters": {
        "width": 280,
        "height": 180,
        "content": "### Retrieve Message Batch Results\n\n[User guide](https://docs.anthropic.com/en/docs/build-with-claude/batch-processing)"
      },
      "typeVersion": 1
    },
    {
      "id": "b87673b1-f08d-4c51-8ee5-4d54557cb382",
      "name": "Sticky Note6",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        900,
        380
      ],
      "parameters": {
        "color": 5,
        "width": 820,
        "height": 340,
        "content": "# Example usage with Chat History Node"
      },
      "typeVersion": 1
    },
    {
      "id": "d6d8ac02-7005-40a1-9950-9517e98e315c",
      "name": "Sticky Note7",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        180,
        720
      ],
      "parameters": {
        "width": 1540,
        "height": 220,
        "content": "# Example usage with single query string"
      },
      "typeVersion": 1
    },
    {
      "id": "0d63deb0-dece-4502-9020-d67c1f194466",
      "name": "Sticky Note8",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        180,
        320
      ],
      "parameters": {
        "color": 3,
        "width": 660,
        "height": 400,
        "content": "# Environment setup\nFor Chat History Node"
      },
      "typeVersion": 1
    },
    {
      "id": "cab94e09-6b84-4a38-b854-670241744db5",
      "name": "Sticky Note9",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        2120,
        800
      ],
      "parameters": {
        "height": 220,
        "content": "## anthropic-version\n\n[Documentation](https://docs.anthropic.com/en/api/versioning)\n\nWhen making API requests, you must send an anthropic-version request header. For example, anthropic-version: `2023-06-01` (latest supported version)"
      },
      "typeVersion": 1
    },
    {
      "id": "ab0a51a1-3c84-4a88-968b-fd46ab07de85",
      "name": "Sticky Note10",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        2560,
        400
      ],
      "parameters": {
        "color": 5,
        "width": 480,
        "height": 300,
        "content": "# Example usage with Chat History Node (result)"
      },
      "typeVersion": 1
    },
    {
      "id": "d91b9be7-ef32-48d6-b880-cab0e99ba9bc",
      "name": "Sticky Note11",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        2560,
        700
      ],
      "parameters": {
        "width": 480,
        "height": 300,
        "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n# Example usage with single query string (result)"
      },
      "typeVersion": 1
    },
    {
      "id": "341811e9-6677-42d9-be28-c388dbf68101",
      "name": "Join two example requests into array",
      "type": "n8n-nodes-base.merge",
      "position": [
        1748,
        625
      ],
      "parameters": {},
      "typeVersion": 3.1
    },
    {
      "id": "45a09f05-7610-4b0a-ab7f-0094c4b3f318",
      "name": "Append execution data for single query example",
      "type": "n8n-nodes-base.set",
      "notes": "custom_id, model and max tokens",
      "position": [
        1010,
        775
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "8276602f-689f-45c2-bce0-5df8500912b6",
              "name": "custom_id",
              "type": "string",
              "value": "second-prompt-in-my-batch"
            },
            {
              "id": "2c513dc2-d8cb-4ba3-b3c1-ea79517b9434",
              "name": "model",
              "type": "string",
              "value": "claude-3-5-haiku-20241022"
            },
            {
              "id": "b052140b-1152-4327-9c5a-5030b78990b7",
              "name": "max_tokens",
              "type": "number",
              "value": 100
            }
          ]
        },
        "includeOtherFields": true
      },
      "notesInFlow": true,
      "typeVersion": 3.4
    },
    {
      "id": "c4e35349-840c-4c81-852c-0d8cd9331364",
      "name": "Append execution data for chat memory example",
      "type": "n8n-nodes-base.set",
      "notes": "custom_id, model and max tokens",
      "position": [
        1308,
        475
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "8276602f-689f-45c2-bce0-5df8500912b6",
              "name": "custom_id",
              "type": "string",
              "value": "first-prompt-in-my-batch"
            },
            {
              "id": "2c513dc2-d8cb-4ba3-b3c1-ea79517b9434",
              "name": "model",
              "type": "string",
              "value": "claude-3-5-haiku-20241022"
            },
            {
              "id": "b052140b-1152-4327-9c5a-5030b78990b7",
              "name": "max_tokens",
              "type": "number",
              "value": 100
            }
          ]
        },
        "includeOtherFields": true
      },
      "notesInFlow": true,
      "typeVersion": 3.4
    },
    {
      "id": "058aedb1-fdfe-4edc-8d51-3b93ec7d232d",
      "name": "Truncate Chat Memory",
      "type": "@n8n/n8n-nodes-langchain.memoryManager",
      "notes": "ensure clean state",
      "position": [
        180,
        475
      ],
      "parameters": {
        "mode": "delete",
        "deleteMode": "all"
      },
      "notesInFlow": true,
      "typeVersion": 1.1
    }
  ],
  "pinData": {},
  "connections": {
    "Get results": {
      "main": [
        [
          {
            "node": "Parse response",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Run example": {
      "main": [
        [
          {
            "node": "One query example",
            "type": "main",
            "index": 0
          },
          {
            "node": "Truncate Chat Memory",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Submit batch": {
      "main": [
        [
          {
            "node": "If ended processing",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Parse response": {
      "main": [
        [
          {
            "node": "Split Out Parsed Results",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "One query example": {
      "main": [
        [
          {
            "node": "Append execution data for single query example",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Check batch status": {
      "main": [
        [
          {
            "node": "If ended processing",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "If ended processing": {
      "main": [
        [
          {
            "node": "Get results",
            "type": "main",
            "index": 0
          }
        ],
        [
          {
            "node": "Batch Status Poll Interval",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Simple Memory Store": {
      "ai_memory": [
        [
          {
            "node": "Load Chat Memory Data",
            "type": "ai_memory",
            "index": 0
          },
          {
            "node": "Fill Chat Memory with example data",
            "type": "ai_memory",
            "index": 0
          },
          {
            "node": "Truncate Chat Memory",
            "type": "ai_memory",
            "index": 0
          }
        ]
      ]
    },
    "Truncate Chat Memory": {
      "main": [
        [
          {
            "node": "Fill Chat Memory with example data",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Load Chat Memory Data": {
      "main": [
        [
          {
            "node": "Append execution data for chat memory example",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Batch Status Poll Interval": {
      "main": [
        [
          {
            "node": "Check batch status",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Construct 'requests' array": {
      "main": [
        [
          {
            "node": "Set desired 'anthropic-version'",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Delete original properties": {
      "main": [
        [
          {
            "node": "Join two example requests into array",
            "type": "main",
            "index": 1
          }
        ]
      ]
    },
    "Filter First Prompt Results": {
      "main": [
        [
          {
            "node": "First Prompt Result",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Filter Second Prompt Results": {
      "main": [
        [
          {
            "node": "Second Prompt Result",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Set desired 'anthropic-version'": {
      "main": [
        [
          {
            "node": "Execute Workflow 'Process Multiple Prompts in Parallel with Anthropic Claude Batch API'",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "When Executed by Another Workflow": {
      "main": [
        [
          {
            "node": "Submit batch",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Fill Chat Memory with example data": {
      "main": [
        [
          {
            "node": "Load Chat Memory Data",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Join two example requests into array": {
      "main": [
        [
          {
            "node": "Construct 'requests' array",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Append execution data for chat memory example": {
      "main": [
        [
          {
            "node": "Build batch 'request' object from Chat Memory and execution data",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Build batch 'request' object for single query": {
      "main": [
        [
          {
            "node": "Delete original properties",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Append execution data for single query example": {
      "main": [
        [
          {
            "node": "Build batch 'request' object for single query",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Build batch 'request' object from Chat Memory and execution data": {
      "main": [
        [
          {
            "node": "Join two example requests into array",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Execute Workflow 'Process Multiple Prompts in Parallel with Anthropic Claude Batch API'": {
      "main": [
        [
          {
            "node": "Filter First Prompt Results",
            "type": "main",
            "index": 0
          },
          {
            "node": "Filter Second Prompt Results",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  }
}

Workflow n8n gestion de données : pour qui est ce workflow ?

Ce workflow s'adresse principalement aux équipes techniques, telles que les développeurs et les data scientists, qui travaillent avec des API. Il est idéal pour les entreprises de taille moyenne à grande cherchant à automatiser leurs processus de gestion de données sans nécessiter de compétences avancées en programmation.

Workflow n8n gestion de données : problème résolu

Ce workflow résout le problème de la gestion manuelle des requêtes API, souvent source d'erreurs et de pertes de temps. En automatisant le processus de soumission, de vérification et de récupération des résultats, il permet aux utilisateurs de gagner en efficacité et en précision. Après mise en place, les utilisateurs peuvent traiter des requêtes en batch de manière fluide, réduisant ainsi le risque d'erreurs et améliorant la réactivité de leurs systèmes.

Workflow n8n gestion de données : étapes du workflow

Étape 1 : Le workflow est déclenché manuellement.

  • Étape 1 : Soumission d'un lot de requêtes via le nœud 'Submit batch'.
  • Étape 2 : Vérification de l'état des requêtes avec 'Check batch status'.
  • Étape 3 : Récupération des résultats grâce au nœud 'Get results'.
  • Étape 4 : Analyse des résultats avec 'Parse response'.
  • Étape 5 : Gestion des flux conditionnels avec 'If ended processing'.
  • Étape 6 : Utilisation de la mémoire pour stocker et gérer les données tout au long du processus.

Workflow n8n gestion de données : guide de personnalisation

Pour personnaliser ce workflow, vous pouvez modifier l'URL de l'API dans le nœud 'Submit batch' pour l'adapter à votre service. Assurez-vous également de configurer les paramètres d'authentification nécessaires pour accéder à l'API. Si vous souhaitez ajouter des fonctionnalités, envisagez d'intégrer d'autres nœuds n8n pour enrichir le traitement des données. Pour sécuriser le flux, vérifiez les permissions d'accès et mettez en place des notifications en cas d'erreurs dans le traitement des requêtes.