{ "name": "AI Use Case Intake Classifier", "nodes": [ { "parameters": { "httpMethod": "POST", "path": "ai-intake", "responseMode": "responseNode", "options": {} }, "id": "webhook-trigger", "name": "Intake Form Webhook", "type": "n8n-nodes-base.webhook", "typeVersion": 1, "position": [240, 300], "webhookId": "ai-intake-classifier" }, { "parameters": { "jsCode": "// Validate and extract intake form fields\nconst body = $input.first().json.body || $input.first().json;\n\nconst required = ['use_case_title', 'business_unit', 'problem_statement', 'data_sources', 'expected_outcome'];\nconst missing = required.filter(field => !body[field]);\n\nif (missing.length > 0) {\n throw new Error(`Missing required fields: ${missing.join(', ')}`);\n}\n\n// Build structured intake object\nconst intake = {\n id: `INTAKE-${Date.now()}`,\n timestamp: new Date().toISOString(),\n submitter_email: body.submitter_email || 'unknown@company.com',\n use_case_title: body.use_case_title,\n business_unit: body.business_unit,\n problem_statement: body.problem_statement,\n current_process: body.current_process || 'Not provided',\n data_sources: body.data_sources,\n data_sensitivity: body.data_sensitivity || 'unknown',\n expected_outcome: body.expected_outcome,\n estimated_users: body.estimated_users || 'unknown',\n estimated_roi: body.estimated_roi || 'Not provided',\n external_vendor: body.external_vendor || false,\n regulatory_domain: body.regulatory_domain || 'none'\n};\n\nreturn [{ json: intake }];" }, "id": "validate-extract", "name": "Validate & Extract Fields", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [460, 300] }, { "parameters": { "authentication": "genericCredentialType", "genericAuthType": "httpHeaderAuth", "url": "https://api.anthropic.com/v1/messages", "sendHeaders": true, "headerParameters": { "parameters": [ { "name": "x-api-key", "value": "={{$credentials.httpHeaderAuth.value}}" }, { "name": "anthropic-version", "value": "2023-06-01" }, { "name": "content-type", "value": "application/json" } ] }, "sendBody": true, "bodyParameters": { "parameters": [] }, "options": {}, "specifyBody": "json", "jsonBody": "={\n \"model\": \"claude-3-haiku-20240307\",\n \"max_tokens\": 1024,\n \"messages\": [\n {\n \"role\": \"user\",\n \"content\": \"You are an Enterprise AI Governance specialist. Classify the following AI use case intake request into a risk tier and provide a structured assessment.\\n\\nRISK TIER DEFINITIONS:\\n- LOW: Internal productivity tools, no sensitive data, limited automation, reversible actions. Auto-approved.\\n- MEDIUM: Involves internal business data, moderate automation, some human review required. Requires manager approval.\\n- HIGH: Sensitive customer/employee data, significant automation, regulatory implications, or external vendor AI. Requires AI Governance Board review.\\n- CRITICAL: PII at scale, financial decisions, healthcare data, legal implications, or fully autonomous decisions with irreversible consequences. Requires CISO + Legal + AI Governance Board.\\n\\nINTAKE REQUEST:\\nTitle: {{ $json.use_case_title }}\\nBusiness Unit: {{ $json.business_unit }}\\nProblem: {{ $json.problem_statement }}\\nCurrent Process: {{ $json.current_process }}\\nData Sources: {{ $json.data_sources }}\\nData Sensitivity: {{ $json.data_sensitivity }}\\nExpected Outcome: {{ $json.expected_outcome }}\\nEstimated Users: {{ $json.estimated_users }}\\nEstimated ROI: {{ $json.estimated_roi }}\\nExternal Vendor AI: {{ $json.external_vendor }}\\nRegulatory Domain: {{ $json.regulatory_domain }}\\n\\nRespond ONLY with valid JSON in this exact format:\\n{\\n \\\"risk_tier\\\": \\\"LOW|MEDIUM|HIGH|CRITICAL\\\",\\n \\\"risk_score\\\": 1-10,\\n \\\"primary_risk_factors\\\": [\\\"factor1\\\", \\\"factor2\\\"],\\n \\\"required_controls\\\": [\\\"control1\\\", \\\"control2\\\"],\\n \\\"approval_path\\\": \\\"Auto-approved|Manager|AI Governance Board|CISO + Legal + AI Governance Board\\\",\\n \\\"estimated_review_days\\\": 0-30,\\n \\\"classification_rationale\\\": \\\"2-3 sentence explanation\\\",\\n \\\"recommended_next_steps\\\": [\\\"step1\\\", \\\"step2\\\"]\\n}\"\n }\n ]\n}" }, "id": "claude-classifier", "name": "Claude Risk Classifier", "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.1, "position": [680, 300] }, { "parameters": { "jsCode": "// Parse Claude's classification response\nconst claudeResponse = $input.first().json;\nconst intake = $('Validate & Extract Fields').first().json;\n\n// Extract text content from Claude response\nconst content = claudeResponse.content[0].text;\n\nlet classification;\ntry {\n classification = JSON.parse(content);\n} catch (e) {\n // Fallback if Claude returns non-JSON\n classification = {\n risk_tier: 'MEDIUM',\n risk_score: 5,\n primary_risk_factors: ['Classification parsing error - manual review required'],\n required_controls: ['Manual risk assessment'],\n approval_path: 'AI Governance Board',\n estimated_review_days: 5,\n classification_rationale: 'Automated classification failed. Manual review triggered.',\n recommended_next_steps: ['Contact AI Governance team for manual review']\n };\n}\n\n// Merge intake + classification\nconst result = {\n ...intake,\n classification: classification,\n status: 'CLASSIFIED',\n classified_at: new Date().toISOString()\n};\n\nreturn [{ json: result }];" }, "id": "parse-classification", "name": "Parse Classification", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [900, 300] }, { "parameters": { "conditions": { "string": [ { "value1": "={{ $json.classification.risk_tier }}", "operation": "equals", "value2": "LOW" } ] } }, "id": "route-low", "name": "Is LOW Risk?", "type": "n8n-nodes-base.if", "typeVersion": 1, "position": [1120, 180] }, { "parameters": { "conditions": { "string": [ { "value1": "={{ $json.classification.risk_tier }}", "operation": "equals", "value2": "MEDIUM" } ] } }, "id": "route-medium", "name": "Is MEDIUM Risk?", "type": "n8n-nodes-base.if", "typeVersion": 1, "position": [1120, 380] }, { "parameters": { "conditions": { "string": [ { "value1": "={{ $json.classification.risk_tier }}", "operation": "equals", "value2": "HIGH" } ] } }, "id": "route-high", "name": "Is HIGH Risk?", "type": "n8n-nodes-base.if", "typeVersion": 1, "position": [1120, 560] }, { "parameters": { "jsCode": "// AUTO-APPROVE: LOW risk use cases\nconst data = $input.first().json;\n\nreturn [{\n json: {\n ...data,\n approval_decision: 'AUTO_APPROVED',\n approved_by: 'AI Governance System',\n approval_timestamp: new Date().toISOString(),\n notification_sent_to: data.submitter_email,\n next_action: 'Submit to AI Platform team for build queue',\n message: `Your AI use case \"${data.use_case_title}\" has been automatically approved (LOW risk). It has been added to the AI Platform build queue. Expected review: ${data.classification.estimated_review_days} business days.`\n }\n}];" }, "id": "auto-approve", "name": "Auto-Approve LOW", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [1380, 120] }, { "parameters": { "jsCode": "// ROUTE TO MANAGER: MEDIUM risk use cases\nconst data = $input.first().json;\n\n// Determine approver based on business unit\nconst approverMap = {\n 'Sales': 'sales-manager@company.com',\n 'HR': 'hr-manager@company.com',\n 'Finance': 'finance-manager@company.com',\n 'Legal': 'legal-manager@company.com',\n 'Engineering': 'eng-manager@company.com',\n 'Marketing': 'marketing-manager@company.com',\n 'Operations': 'ops-manager@company.com'\n};\n\nconst approver = approverMap[data.business_unit] || 'ai-governance@company.com';\n\nreturn [{\n json: {\n ...data,\n approval_decision: 'PENDING_MANAGER_APPROVAL',\n assigned_approver: approver,\n approval_deadline: new Date(Date.now() + 3 * 24 * 60 * 60 * 1000).toISOString(),\n notification_sent_to: [data.submitter_email, approver],\n next_action: `Awaiting manager approval from ${approver}`,\n message: `Your AI use case \"${data.use_case_title}\" requires manager approval (MEDIUM risk). Your manager at ${approver} has been notified. Expected decision: ${data.classification.estimated_review_days} business days.`\n }\n}];" }, "id": "route-to-manager", "name": "Route to Manager", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [1380, 340] }, { "parameters": { "jsCode": "// ROUTE TO AI GOVERNANCE BOARD: HIGH risk\nconst data = $input.first().json;\n\nreturn [{\n json: {\n ...data,\n approval_decision: 'PENDING_GOVERNANCE_BOARD',\n assigned_approver: 'ai-governance-board@company.com',\n approval_deadline: new Date(Date.now() + 10 * 24 * 60 * 60 * 1000).toISOString(),\n notification_sent_to: [data.submitter_email, 'ai-governance-board@company.com'],\n required_documentation: [\n 'Data Privacy Impact Assessment',\n 'Security Risk Assessment',\n 'Bias & Fairness Evaluation Plan',\n 'Human-in-the-loop documentation',\n 'Audit trail specification'\n ],\n next_action: 'Complete required documentation and submit to AI Governance Board',\n message: `Your AI use case \"${data.use_case_title}\" requires AI Governance Board review (HIGH risk). Complete the required documentation within 5 business days. Board review: ${data.classification.estimated_review_days} business days.`\n }\n}];" }, "id": "route-to-board", "name": "Route to Governance Board", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [1380, 520] }, { "parameters": { "jsCode": "// ESCALATE: CRITICAL risk — CISO + Legal + Governance Board\nconst data = $input.first().json;\n\nreturn [{\n json: {\n ...data,\n approval_decision: 'CRITICAL_ESCALATION',\n assigned_approver: 'CISO + Legal + AI Governance Board',\n notification_sent_to: [\n data.submitter_email,\n 'ciso@company.com',\n 'legal@company.com',\n 'ai-governance-board@company.com',\n 'cto@company.com'\n ],\n required_documentation: [\n 'Full Data Privacy Impact Assessment (DPIA)',\n 'Legal Compliance Review',\n 'Security Architecture Review',\n 'Executive Risk Acceptance',\n 'Regulatory Compliance Documentation',\n 'Incident Response Plan',\n 'Model Explainability Documentation',\n 'Third-party Vendor Security Assessment (if applicable)'\n ],\n hold_reason: 'Use case placed on HOLD pending CRITICAL risk review',\n next_action: 'CISO + Legal + AI Governance Board notified. Do not proceed without written approval.',\n message: `CRITICAL ALERT: AI use case \"${data.use_case_title}\" has been flagged as CRITICAL risk. CISO, Legal, and AI Governance Board have been notified. Use case is on HOLD. Do not proceed without written executive approval.`\n }\n}];" }, "id": "critical-escalation", "name": "Critical Escalation", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [1380, 700] }, { "parameters": { "jsCode": "// LOG to audit trail — all risk levels\nconst data = $input.first().json;\n\nconst auditEntry = {\n audit_id: `AUDIT-${Date.now()}`,\n intake_id: data.id,\n timestamp: new Date().toISOString(),\n use_case_title: data.use_case_title,\n business_unit: data.business_unit,\n submitter: data.submitter_email,\n risk_tier: data.classification.risk_tier,\n risk_score: data.classification.risk_score,\n approval_decision: data.approval_decision,\n assigned_approver: data.assigned_approver || 'System Auto-Approved',\n primary_risk_factors: data.classification.primary_risk_factors,\n required_controls: data.classification.required_controls,\n classification_rationale: data.classification.classification_rationale,\n action_taken: `Routed to: ${data.approval_decision}`\n};\n\nconsole.log('AUDIT LOG:', JSON.stringify(auditEntry, null, 2));\n\nreturn [{ json: { ...data, audit_entry: auditEntry } }];" }, "id": "audit-log", "name": "Audit Log", "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [1620, 400] }, { "parameters": { "respondWith": "json", "responseBody": "={{ JSON.stringify({\n success: true,\n intake_id: $json.id,\n risk_tier: $json.classification.risk_tier,\n risk_score: $json.classification.risk_score,\n approval_decision: $json.approval_decision,\n assigned_approver: $json.assigned_approver || 'Auto-Approved',\n estimated_review_days: $json.classification.estimated_review_days,\n next_action: $json.next_action,\n message: $json.message,\n required_controls: $json.classification.required_controls,\n primary_risk_factors: $json.classification.primary_risk_factors\n}, null, 2) }}", "options": { "responseHeaders": { "entries": [ { "name": "Content-Type", "value": "application/json" } ] } } }, "id": "send-response", "name": "Send Response", "type": "n8n-nodes-base.respondToWebhook", "typeVersion": 1, "position": [1840, 400] } ], "connections": { "Intake Form Webhook": { "main": [ [ { "node": "Validate & Extract Fields", "type": "main", "index": 0 } ] ] }, "Validate & Extract Fields": { "main": [ [ { "node": "Claude Risk Classifier", "type": "main", "index": 0 } ] ] }, "Claude Risk Classifier": { "main": [ [ { "node": "Parse Classification", "type": "main", "index": 0 } ] ] }, "Parse Classification": { "main": [ [ { "node": "Is LOW Risk?", "type": "main", "index": 0 }, { "node": "Is MEDIUM Risk?", "type": "main", "index": 0 }, { "node": "Is HIGH Risk?", "type": "main", "index": 0 } ] ] }, "Is LOW Risk?": { "main": [ [ { "node": "Auto-Approve LOW", "type": "main", "index": 0 } ] ] }, "Is MEDIUM Risk?": { "main": [ [ { "node": "Route to Manager", "type": "main", "index": 0 } ] ] }, "Is HIGH Risk?": { "main": [ [ { "node": "Route to Governance Board", "type": "main", "index": 0 } ], [ { "node": "Critical Escalation", "type": "main", "index": 0 } ] ] }, "Auto-Approve LOW": { "main": [ [ { "node": "Audit Log", "type": "main", "index": 0 } ] ] }, "Route to Manager": { "main": [ [ { "node": "Audit Log", "type": "main", "index": 0 } ] ] }, "Route to Governance Board": { "main": [ [ { "node": "Audit Log", "type": "main", "index": 0 } ] ] }, "Critical Escalation": { "main": [ [ { "node": "Audit Log", "type": "main", "index": 0 } ] ] }, "Audit Log": { "main": [ [ { "node": "Send Response", "type": "main", "index": 0 } ] ] } }, "settings": { "executionOrder": "v1" }, "staticData": null, "tags": [ { "name": "AI Governance", "createdAt": "2026-05-29" }, { "name": "Trust by Design", "createdAt": "2026-05-29" } ], "meta": { "templateCredsSetupCompleted": false }, "pinData": {}, "versionId": "1.0.0" }