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": "Multi-Agent Research Pipeline",
"nodes": [
{
"parameters": {
"httpMethod": "POST",
"path": "research",
"responseMode": "responseNode",
"options": {
"rawBody": false
}
},
"id": "n8n-webhook-trigger",
"name": "Webhook Trigger",
"type": "n8n-nodes-base.webhook",
"typeVersion": 1,
"position": [
240,
300
],
"notes": "Entry point: receives POST /webhook/research with {query, search_depth, metadata}"
},
{
"parameters": {
"jsCode": "// Validate and normalize the incoming research request\nconst body = $input.first().json.body || $input.first().json;\n\nif (!body.query || body.query.trim().length < 3) {\n throw new Error('Invalid query: must be at least 3 characters');\n}\n\nconst normalized = {\n query: body.query.trim(),\n search_depth: body.search_depth || 'advanced',\n use_cache: body.use_cache !== false,\n notify_n8n: false, // Avoid loop: don't re-notify n8n from pipeline\n metadata: {\n source: body.metadata?.source || 'n8n',\n user_id: body.metadata?.user_id || 'anonymous',\n timestamp: new Date().toISOString()\n }\n};\n\nreturn [{ json: normalized }];"
},
"id": "n8n-validate-input",
"name": "Validate & Normalize Input",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
460,
300
],
"notes": "Validates required fields, sets defaults, prevents webhook loop"
},
{
"parameters": {
"method": "POST",
"url": "={{ $env.RESEARCH_API_URL }}/api/research/",
"sendHeaders": true,
"headerParameters": {
"parameters": [
{
"name": "Content-Type",
"value": "application/json"
},
{
"name": "X-API-Key",
"value": "={{ $env.RESEARCH_API_SECRET }}"
}
]
},
"sendBody": true,
"bodyParameters": {
"parameters": [
{
"name": "query",
"value": "={{ $json.query }}"
},
{
"name": "search_depth",
"value": "={{ $json.search_depth }}"
},
{
"name": "use_cache",
"value": "={{ $json.use_cache }}"
},
{
"name": "notify_n8n",
"value": false
}
]
},
"options": {
"timeout": 120000,
"response": {
"response": {
"neverError": false
}
}
}
},
"id": "n8n-call-pipeline",
"name": "Call Research Pipeline",
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4,
"position": [
680,
300
],
"notes": "Calls FastAPI research endpoint. Timeout: 120s to allow full pipeline run"
},
{
"parameters": {
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict"
},
"conditions": [
{
"id": "check-status",
"leftValue": "={{ $json.status }}",
"rightValue": "completed",
"operator": {
"type": "string",
"operation": "equals"
}
}
],
"combinator": "and"
},
"options": {}
},
"id": "n8n-check-status",
"name": "Check Pipeline Status",
"type": "n8n-nodes-base.if",
"typeVersion": 2,
"position": [
900,
300
],
"notes": "Routes: completed \u2192 format report | partial/failed \u2192 error handler"
},
{
"parameters": {
"jsCode": "// Format the research report into a clean output document\nconst report = $input.first().json;\n\nconst verdict = report.fact_check?.verdict || 'UNKNOWN';\nconst confidence = Math.round((report.overall_confidence || 0) * 100);\n\nconst sourceList = (report.sources || [])\n .slice(0, 5)\n .map((s, i) => `${i + 1}. [${s.title}](${s.url}) \u2014 Score: ${(s.score * 100).toFixed(0)}%`)\n .join('\\n');\n\nconst keyPoints = (report.summary?.key_points || [])\n .map((p, i) => `\u2022 ${p}`)\n .join('\\n');\n\nconst formatted = {\n // Metadata\n query: report.query,\n generated_at: report.completed_at,\n latency_seconds: Math.round(report.total_latency_ms / 1000),\n overall_confidence_pct: confidence,\n verdict: verdict,\n verdict_emoji: report.fact_check?.verdict_emoji || '\u2753',\n\n // Formatted text sections\n executive_summary: report.summary?.summary || 'No summary available.',\n key_findings: keyPoints,\n top_sources: sourceList,\n fact_check_assessment: report.fact_check?.fact_check_summary || '',\n\n // Structured data\n key_entities: report.summary?.key_entities || {},\n contradictions: report.fact_check?.contradictions || [],\n unverified_claims: report.fact_check?.unverified_claims || [],\n\n // Full report passthrough\n _raw: report\n};\n\nreturn [{ json: formatted }];"
},
"id": "n8n-format-report",
"name": "Format Research Report",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
1120,
200
],
"notes": "Transforms raw pipeline output into formatted report with confidence % and verdict"
},
{
"parameters": {
"jsCode": "// Generate a Markdown-formatted report document\nconst r = $input.first().json;\n\nconst md = `# Research Report\n\n**Query:** ${r.query}\n**Generated:** ${r.generated_at}\n**Confidence:** ${r.overall_confidence_pct}% ${r.verdict_emoji}\n**Verdict:** ${r.verdict}\n**Pipeline Time:** ${r.latency_seconds}s\n\n---\n\n## Executive Summary\n\n${r.executive_summary}\n\n## Key Findings\n\n${r.key_findings}\n\n## Sources\n\n${r.top_sources}\n\n## Fact-Check Assessment\n\n${r.fact_check_assessment || '_No fact-check data available._'}\n\n${r.contradictions.length > 0 ? '## \u26a0\ufe0f Contradictions Found\\n\\n' + r.contradictions.map(c => '- ' + c).join('\\n') : ''}\n\n---\n*Generated by Multi-Agent Research Assistant*\n`;\n\nreturn [{ json: { ...r, markdown_report: md } }];"
},
"id": "n8n-generate-markdown",
"name": "Generate Markdown Report",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
1340,
200
],
"notes": "Generates a Markdown document suitable for Notion, Google Docs, Slack, or email"
},
{
"parameters": {
"respondWith": "json",
"responseBody": "={{ JSON.stringify($json, null, 2) }}",
"options": {
"responseCode": 200,
"responseHeaders": {
"entries": [
{
"name": "Content-Type",
"value": "application/json"
}
]
}
}
},
"id": "n8n-respond-success",
"name": "Return Success Response",
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1,
"position": [
1560,
200
],
"notes": "Returns the formatted report to the caller"
},
{
"parameters": {
"jsCode": "const report = $input.first().json;\nconst errors = report.errors || [];\n\nreturn [{\n json: {\n error: true,\n status: report.status || 'failed',\n query: report.query,\n errors: errors,\n message: errors.length > 0 ? errors.join('; ') : 'Pipeline returned partial results',\n partial_data: {\n sources_count: (report.sources || []).length,\n has_summary: !!report.summary,\n has_fact_check: !!report.fact_check\n }\n }\n}];"
},
"id": "n8n-handle-error",
"name": "Handle Partial/Error",
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
1120,
420
],
"notes": "Handles partial results or failed pipeline runs"
},
{
"parameters": {
"respondWith": "json",
"responseBody": "={{ JSON.stringify($json, null, 2) }}",
"options": {
"responseCode": 206,
"responseHeaders": {
"entries": [
{
"name": "Content-Type",
"value": "application/json"
}
]
}
}
},
"id": "n8n-respond-partial",
"name": "Return Partial Response",
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1,
"position": [
1340,
420
],
"notes": "Returns 206 Partial Content for incomplete pipeline runs"
}
],
"connections": {
"Webhook Trigger": {
"main": [
[
{
"node": "Validate & Normalize Input",
"type": "main",
"index": 0
}
]
]
},
"Validate & Normalize Input": {
"main": [
[
{
"node": "Call Research Pipeline",
"type": "main",
"index": 0
}
]
]
},
"Call Research Pipeline": {
"main": [
[
{
"node": "Check Pipeline Status",
"type": "main",
"index": 0
}
]
]
},
"Check Pipeline Status": {
"main": [
[
{
"node": "Format Research Report",
"type": "main",
"index": 0
}
],
[
{
"node": "Handle Partial/Error",
"type": "main",
"index": 0
}
]
]
},
"Format Research Report": {
"main": [
[
{
"node": "Generate Markdown Report",
"type": "main",
"index": 0
}
]
]
},
"Generate Markdown Report": {
"main": [
[
{
"node": "Return Success Response",
"type": "main",
"index": 0
}
]
]
},
"Handle Partial/Error": {
"main": [
[
{
"node": "Return Partial Response",
"type": "main",
"index": 0
}
]
]
}
},
"settings": {
"executionOrder": "v1",
"saveManualExecutions": true,
"callerPolicy": "workflowsFromSameOwner",
"errorWorkflow": ""
},
"staticData": null,
"meta": {
"templateCredsSetupCompleted": true
},
"versionId": "1.0.0",
"triggerCount": 1,
"tags": [
{
"id": "research",
"name": "research"
},
{
"id": "ai-pipeline",
"name": "ai-pipeline"
},
{
"id": "multi-agent",
"name": "multi-agent"
}
]
}
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
Multi-Agent Research Pipeline. Uses httpRequest. Webhook trigger; 9 nodes.
Source: https://github.com/palpriyanshu94/Multi-Agent-Research-Assistant/blob/cad41af47ae994f2893ba55ac9da62323d7b0e76/n8n/workflow.json — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
Jigsaw API key for image processing, I use this as a gatekeeper/second pair of eyes. LINK to their website https://jigsawstack.com/ SECOND A postgress DATABASE (I use Supabase) LlamaCloud for the pars
HTTP Request, Postgres, Stop And Error +2
Whatsapp Multi Agent System optimized copy 2.0. Uses airtable, httpRequest, errorTrigger. Webhook trigger; 44 nodes.
Airtable, HTTP Request, Error Trigger
Invoice Agent. Uses httpRequest, emailSend. Webhook trigger; 29 nodes.
HTTP Request, Email Send
Reputation Engine — SEO QA Agent. Uses httpRequest. Webhook trigger; 28 nodes.
HTTP Request
This workflow handles incoming voice calls or audio messages, transcribes them using Whisper (OpenAI) or ElevenLabs, extracts booking intent and preferred time slots using AI, checks availability on C
HTTP Request