Trigger CrewAI Agent via Webhook

The workflow JSON

Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →

{
  "name": "\ud83e\udd16 Trigger CrewAI Agent",
  "nodes": [
    {
      "parameters": {
        "httpMethod": "POST",
        "path": "trigger-agent",
        "responseMode": "responseNode",
        "options": {}
      },
      "name": "Webhook Trigger",
      "type": "n8n-nodes-base.webhook",
      "typeVersion": 1,
      "position": [
        300,
        300
      ]
    },
    {
      "parameters": {
        "requestMethod": "POST",
        "url": "http://host.docker.internal:8010/crews/research",
        "jsonParameters": true,
        "options": {
          "timeout": 30000
        },
        "bodyParametersJson": "={{ JSON.stringify({ \"topic\": $json.body.topic || \"AI trends in 2024\", \"output_type\": $json.body.output_type || \"report\", \"llm_provider\": $json.body.llm_provider || null, \"llm_model\": $json.body.llm_model || null }) }}"
      },
      "name": "Start Research Crew",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 2,
      "position": [
        550,
        300
      ]
    },
    {
      "parameters": {
        "amount": 5,
        "unit": "seconds"
      },
      "name": "Wait",
      "type": "n8n-nodes-base.wait",
      "typeVersion": 1,
      "position": [
        800,
        300
      ]
    },
    {
      "parameters": {
        "requestMethod": "GET",
        "url": "=http://host.docker.internal:8010/jobs/{{ $node['Start Research Crew'].json.job_id }}",
        "options": {
          "timeout": 10000
        }
      },
      "name": "Check Job Status",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 2,
      "position": [
        1050,
        300
      ]
    },
    {
      "parameters": {
        "conditions": {
          "string": [
            {
              "value1": "={{ $json.status }}",
              "operation": "equals",
              "value2": "completed"
            }
          ]
        }
      },
      "name": "Is Completed?",
      "type": "n8n-nodes-base.if",
      "typeVersion": 1,
      "position": [
        1300,
        300
      ]
    },
    {
      "parameters": {
        "conditions": {
          "string": [
            {
              "value1": "={{ $json.status }}",
              "operation": "equals",
              "value2": "failed"
            }
          ]
        }
      },
      "name": "Is Failed?",
      "type": "n8n-nodes-base.if",
      "typeVersion": 1,
      "position": [
        1300,
        500
      ]
    },
    {
      "parameters": {
        "respondWith": "json",
        "responseBody": "={{ JSON.stringify({ \"success\": true, \"job_id\": $node['Start Research Crew'].json.job_id, \"result\": $json.result }) }}"
      },
      "name": "Return Success",
      "type": "n8n-nodes-base.respondToWebhook",
      "typeVersion": 1,
      "position": [
        1550,
        200
      ]
    },
    {
      "parameters": {
        "respondWith": "json",
        "responseBody": "={{ JSON.stringify({ \"success\": false, \"job_id\": $node['Start Research Crew'].json.job_id, \"error\": $json.error }) }}",
        "options": {
          "responseCode": 500
        }
      },
      "name": "Return Error",
      "type": "n8n-nodes-base.respondToWebhook",
      "typeVersion": 1,
      "position": [
        1550,
        500
      ]
    },
    {
      "parameters": {
        "loopCount": 60,
        "options": {}
      },
      "name": "Loop",
      "type": "n8n-nodes-base.splitInBatches",
      "typeVersion": 1,
      "position": [
        1550,
        700
      ]
    },
    {
      "parameters": {
        "amount": 10,
        "unit": "seconds"
      },
      "name": "Wait More",
      "type": "n8n-nodes-base.wait",
      "typeVersion": 1,
      "position": [
        1800,
        700
      ]
    },
    {
      "parameters": {
        "requestMethod": "GET",
        "url": "=http://host.docker.internal:8010/jobs/{{ $node['Start Research Crew'].json.job_id }}",
        "options": {
          "timeout": 10000
        }
      },
      "name": "Poll Job Status",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 2,
      "position": [
        2050,
        700
      ]
    },
    {
      "parameters": {
        "conditions": {
          "string": [
            {
              "value1": "={{ $json.status }}",
              "operation": "equals",
              "value2": "completed"
            }
          ]
        }
      },
      "name": "Poll Completed?",
      "type": "n8n-nodes-base.if",
      "typeVersion": 1,
      "position": [
        2300,
        700
      ]
    },
    {
      "parameters": {
        "respondWith": "json",
        "responseBody": "={{ JSON.stringify({ \"success\": true, \"job_id\": $node['Start Research Crew'].json.job_id, \"result\": $json.result }) }}"
      },
      "name": "Return Poll Success",
      "type": "n8n-nodes-base.respondToWebhook",
      "typeVersion": 1,
      "position": [
        2550,
        600
      ]
    },
    {
      "parameters": {
        "conditions": {
          "string": [
            {
              "value1": "={{ $json.status }}",
              "operation": "equals",
              "value2": "failed"
            }
          ]
        }
      },
      "name": "Poll Failed?",
      "type": "n8n-nodes-base.if",
      "typeVersion": 1,
      "position": [
        2550,
        800
      ]
    },
    {
      "parameters": {
        "respondWith": "json",
        "responseBody": "={{ JSON.stringify({ \"success\": false, \"job_id\": $node['Start Research Crew'].json.job_id, \"error\": $json.error }) }}",
        "options": {
          "responseCode": 500
        }
      },
      "name": "Return Poll Error",
      "type": "n8n-nodes-base.respondToWebhook",
      "typeVersion": 1,
      "position": [
        2800,
        750
      ]
    }
  ],
  "connections": {
    "Webhook Trigger": {
      "main": [
        [
          {
            "node": "Start Research Crew"
          }
        ]
      ]
    },
    "Start Research Crew": {
      "main": [
        [
          {
            "node": "Wait"
          }
        ]
      ]
    },
    "Wait": {
      "main": [
        [
          {
            "node": "Check Job Status"
          }
        ]
      ]
    },
    "Check Job Status": {
      "main": [
        [
          {
            "node": "Is Completed?"
          }
        ]
      ]
    },
    "Is Completed?": {
      "main": [
        [
          {
            "node": "Return Success"
          }
        ],
        [
          {
            "node": "Is Failed?"
          }
        ]
      ]
    },
    "Is Failed?": {
      "main": [
        [
          {
            "node": "Return Error"
          }
        ],
        [
          {
            "node": "Loop"
          }
        ]
      ]
    },
    "Loop": {
      "main": [
        [
          {
            "node": "Wait More"
          }
        ]
      ]
    },
    "Wait More": {
      "main": [
        [
          {
            "node": "Poll Job Status"
          }
        ]
      ]
    },
    "Poll Job Status": {
      "main": [
        [
          {
            "node": "Poll Completed?"
          }
        ]
      ]
    },
    "Poll Completed?": {
      "main": [
        [
          {
            "node": "Return Poll Success"
          }
        ],
        [
          {
            "node": "Poll Failed?"
          }
        ]
      ]
    },
    "Poll Failed?": {
      "main": [
        [
          {
            "node": "Return Poll Error"
          }
        ],
        [
          {
            "node": "Loop"
          }
        ]
      ]
    }
  },
  "active": false,
  "settings": {
    "executionTimeout": 600
  }
}

How this works

Automate complex AI-driven tasks effortlessly by triggering a CrewAI agent through a simple webhook, allowing you to offload research, analysis, or decision-making processes without managing the underlying AI infrastructure. This workflow suits developers, product managers, or business analysts who need reliable, scalable automation for repetitive cognitive workloads, such as market research or content generation. The key step involves sending an HTTP request to initiate the CrewAI agent, followed by polling its status until completion, ensuring you receive a seamless response via the webhook integration.

Use this workflow when integrating CrewAI into chatbots, APIs, or real-time applications requiring asynchronous AI processing, particularly for tasks that take minutes to hours. Avoid it for simple, synchronous operations or when direct API calls suffice without status monitoring, as the polling adds unnecessary overhead. Common variations include adding error retries for flaky agents or chaining multiple CrewAI tasks for sequential workflows.

About this workflow

🤖 Trigger CrewAI Agent. Uses httpRequest, respondToWebhook, splitInBatches. Webhook trigger; 15 nodes.

Source: https://github.com/vagulans/ai-platform/blob/965d60c2bd5d7380ee7c94d3f39799d877e0641d/n8n-workflows/03-trigger-agent.json — original creator credit. Request a take-down →

More AI & RAG workflows → · Browse all categories →

Related workflows

Workflows that share integrations, category, or trigger type with this one. All free to copy and import.