{ "edges": [ { "to": "d1a1b1c1-0004-4000-8000-000000000004", "from": "d1a1b1c1-0003-4000-8000-000000000003", "toPort": "0", "fromPort": "0" }, { "to": "d1a1b1c1-0011-4000-8000-000000000011", "from": "d1a1b1c1-0010-4000-8000-000000000010", "toPort": "0", "fromPort": "0" }, { "to": "d1a1b1c1-0002-4000-8000-000000000002", "from": "d1a1b1c1-0001-4000-8000-000000000001", "toPort": "0", "fromPort": "0" }, { "to": "d1a1b1c1-0009-4000-8000-000000000009", "from": "d1a1b1c1-0008-4000-8000-000000000008", "toPort": "0", "fromPort": "0" }, { "to": "d1a1b1c1-0006-4000-8000-000000000006", "from": "d1a1b1c1-0005-4000-8000-000000000005", "toPort": "0", "fromPort": "0" }, { "to": "d1a1b1c1-0012-4000-8000-000000000012", "from": "d1a1b1c1-0005-4000-8000-000000000005", "toPort": "0", "fromPort": "1" }, { "to": "d1a1b1c1-0005-4000-8000-000000000005", "from": "d1a1b1c1-0004-4000-8000-000000000004", "toPort": "0", "fromPort": "0" }, { "to": "d1a1b1c1-0006-4000-8000-000000000006", "from": "d1a1b1c1-0007-4000-8000-000000000007", "toPort": "0", "fromPort": "ai_languageModel" }, { "to": "d1a1b1c1-0010-4000-8000-000000000010", "from": "d1a1b1c1-0009-4000-8000-000000000009", "toPort": "0", "fromPort": "0" }, { "to": "d1a1b1c1-0008-4000-8000-000000000008", "from": "d1a1b1c1-0006-4000-8000-000000000006", "toPort": "0", "fromPort": "0" }, { "to": "d1a1b1c1-0003-4000-8000-000000000003", "from": "d1a1b1c1-0002-4000-8000-000000000002", "toPort": "0", "fromPort": "0" } ], "nodes": [ { "id": "d1a1b1c1-0001-4000-8000-000000000001", "name": "Daily Check", "type": "trigger", "params": { "kind": "n8n-nodes-base.scheduleTrigger", "rule": { "interval": [ { "triggerAtHour": 8 } ] } }, "position": [ 112, 480 ] }, { "id": "d1a1b1c1-0002-4000-8000-000000000002", "name": "Get Pages to Monitor", "type": "integration", "params": { "options": {}, "service": "googleSheets", "operation": "read", "sheetName": { "__rl": true, "mode": "name", "value": "Pages" }, "documentId": { "__rl": true, "mode": "id", "value": "YOUR_GOOGLE_SHEET_ID" } }, "position": [ 336, 480 ] }, { "id": "d1a1b1c1-0003-4000-8000-000000000003", "name": "Fetch Page", "type": "http", "params": { "url": "={{ $json.url }}", "options": {}, "sendHeaders": true, "headerParameters": { "parameters": [ { "name": "User-Agent", "value": "Mozilla/5.0 (compatible; TOSWatcher/1.0)" } ] } }, "position": [ 544, 480 ] }, { "id": "d1a1b1c1-0004-4000-8000-000000000004", "name": "Extract & Compare", "type": "code", "params": { "mode": "runOnceForEachItem", "jsCode": "// Get HTTP response\nconst resp = $input.item.json;\nconst pageData = $('Get Pages to Monitor').item.json;\nconst checkedAt = new Date().toISOString().split('T')[0];\n\n// Detect HTTP errors\nif (resp.error || resp.statusCode >= 400 || resp.status >= 400) {\n return {\n url: pageData.url,\n page_name: pageData.page_name,\n new_content: '', old_content: '',\n old_content_for_ai: '', new_content_for_ai: '',\n changed: false, is_first_run: false,\n error: 'HTTP error: ' + (resp.error?.message || resp.statusCode || resp.status || 'unknown'),\n checked_at: checkedAt\n };\n}\n\n// Get response body\nlet html = '';\nif (typeof resp.data === 'string') html = resp.data;\nelse if (typeof resp.body === 'string') html = resp.body;\nelse if (typeof resp === 'string') html = resp;\nelse html = JSON.stringify(resp);\n\n// Check for Cloudflare/bot challenges\nif (html.includes('Enable JavaScript and cookies') || html.includes('cf-challenge') || html.includes('_cf_chl')) {\n return {\n url: pageData.url,\n page_name: pageData.page_name,\n new_content: '', old_content: '',\n old_content_for_ai: '', new_content_for_ai: '',\n changed: false, is_first_run: false,\n error: 'Page is behind Cloudflare bot protection - use a different URL',\n checked_at: checkedAt\n };\n}\n\n// Check for empty/tiny responses\nif (!html || html.length < 200) {\n return {\n url: pageData.url,\n page_name: pageData.page_name,\n new_content: '', old_content: '',\n old_content_for_ai: '', new_content_for_ai: '',\n changed: false, is_first_run: false,\n error: 'Page returned empty or very short content',\n checked_at: checkedAt\n };\n}\n\n// Strip HTML tags to get plain text\nlet text = html\n .replace(/]*>[\\s\\S]*?<\\/script>/gi, '')\n .replace(/]*>[\\s\\S]*?<\\/style>/gi, '')\n .replace(/<[^>]+>/g, ' ')\n .replace(/ /gi, ' ')\n .replace(/&/gi, '&')\n .replace(/</gi, '<')\n .replace(/>/gi, '>')\n .replace(/&#\\d+;/gi, '')\n .replace(/\\s+/g, ' ')\n .trim()\n .substring(0, 40000);\n\n// Compare with previously stored content\nconst oldContent = (pageData.last_content || '').trim();\nconst isFirstRun = oldContent === '';\nconst changed = !isFirstRun && oldContent !== text;\n\n// Build smart AI context: find WHERE the texts differ\nlet oldForAi = oldContent.substring(0, 10000);\nlet newForAi = text.substring(0, 10000);\n\nif (changed && oldContent.length > 0) {\n // Find first point of difference\n let diffStart = 0;\n const minLen = Math.min(oldContent.length, text.length);\n for (let i = 0; i < minLen; i++) {\n if (oldContent[i] !== text[i]) {\n diffStart = i;\n break;\n }\n if (i === minLen - 1) diffStart = minLen;\n }\n\n // Take context: 500 chars before diff, then 4500 chars after\n const contextBefore = Math.max(0, diffStart - 500);\n const prefix = contextBefore > 0 ? '...[IDENTICAL CONTENT SKIPPED]... ' : '';\n\n const oldDiffSection = prefix + oldContent.substring(contextBefore, contextBefore + 5000);\n const newDiffSection = prefix + text.substring(contextBefore, contextBefore + 5000);\n\n // Also grab the tail end of both texts (last 3000 chars)\n const oldTail = oldContent.length > 5000 ? '\\n\\n...[SKIPPING TO END]...\\n' + oldContent.substring(oldContent.length - 3000) : '';\n const newTail = text.length > 5000 ? '\\n\\n...[SKIPPING TO END]...\\n' + text.substring(text.length - 3000) : '';\n\n oldForAi = (oldDiffSection + oldTail).substring(0, 12000);\n newForAi = (newDiffSection + newTail).substring(0, 12000);\n\n // Add length comparison as context\n oldForAi = '[Total length: ' + oldContent.length + ' chars]\\n' + oldForAi;\n newForAi = '[Total length: ' + text.length + ' chars]\\n' + newForAi;\n}\n\nreturn {\n url: pageData.url,\n page_name: pageData.page_name,\n new_content: text,\n old_content: oldContent,\n old_content_for_ai: oldForAi,\n new_content_for_ai: newForAi,\n changed: changed,\n is_first_run: isFirstRun,\n checked_at: checkedAt\n};" }, "position": [ 768, 480 ] }, { "id": "d1a1b1c1-0005-4000-8000-000000000005", "name": "Content Changed?", "type": "if", "params": { "options": {}, "conditions": { "options": { "version": 2, "leftValue": "", "caseSensitive": true, "typeValidation": "loose" }, "combinator": "and", "conditions": [ { "id": "cond-changed", "operator": { "type": "boolean", "operation": "true", "singleValue": true }, "leftValue": "={{ $json.changed }}", "rightValue": "" } ] } }, "position": [ 992, 480 ] }, { "id": "d1a1b1c1-0006-4000-8000-000000000006", "name": "Summarize Changes", "type": "llm", "params": { "text": "=You are a legal document analyst. You MUST compare the two text versions below word-by-word and identify ONLY real, verifiable differences. Do NOT guess or generalize. Every change you list must quote the actual text that was added, removed, or modified.\n\nPage: {{ $json.page_name }} ({{ $json.url }})\n\n--- OLD VERSION ---\n{{ $json.old_content_for_ai }}\n\n--- NEW VERSION ---\n{{ $json.new_content_for_ai }}\n\nRules:\n- ONLY report changes you can verify by comparing the two texts above\n- Quote the specific text that changed (use \"was: ...\" and \"now: ...\" format)\n- If large sections were added or removed, summarize what was added/removed with a brief quote\n- If you cannot find concrete differences, say so honestly\n- NEVER fabricate or assume changes that are not visible in the text\n\nRespond in this format:\n\n**Overview:** One sentence on what changed.\n\n**Changes:**\n- [Quote the specific old text → new text for each change]\n\n**Impact Level:** LOW (cosmetic/formatting), MEDIUM (policy clarification), or HIGH (new restrictions, rights changes, data handling changes)\n\n**What This Means For Users:** Brief practical impact in 1-2 sentences.", "prompt": "=You are a legal document analyst. You MUST compare the two text versions below word-by-word and identify ONLY real, verifiable differences. Do NOT guess or generalize. Every change you list must quote the actual text that was added, removed, or modified.\n\nPage: {{ $json.page_name }} ({{ $json.url }})\n\n--- OLD VERSION ---\n{{ $json.old_content_for_ai }}\n\n--- NEW VERSION ---\n{{ $json.new_content_for_ai }}\n\nRules:\n- ONLY report changes you can verify by comparing the two texts above\n- Quote the specific text that changed (use \"was: ...\" and \"now: ...\" format)\n- If large sections were added or removed, summarize what was added/removed with a brief quote\n- If you cannot find concrete differences, say so honestly\n- NEVER fabricate or assume changes that are not visible in the text\n\nRespond in this format:\n\n**Overview:** One sentence on what changed.\n\n**Changes:**\n- [Quote the specific old text → new text for each change]\n\n**Impact Level:** LOW (cosmetic/formatting), MEDIUM (policy clarification), or HIGH (new restrictions, rights changes, data handling changes)\n\n**What This Means For Users:** Brief practical impact in 1-2 sentences.", "batching": {}, "provider": "openai", "promptType": "define", "approximated": true }, "position": [ 1216, 272 ] }, { "id": "d1a1b1c1-0007-4000-8000-000000000007", "name": "OpenAI Chat Model", "type": "llm", "params": { "model": "gpt-4o", "options": {}, "provider": "openai", "builtInTools": {} }, "position": [ 1328, 480 ] }, { "id": "d1a1b1c1-0008-4000-8000-000000000008", "name": "Build Email Body", "type": "code", "params": { "jsCode": "// Get AI summaries and build email bodies\nconst items = $input.all();\nconst results = [];\n\nfor (const item of items) {\n const summary = item.json.text || item.json.output || 'No summary available';\n const p = $('Extract & Compare').item.json;\n\n const body = '

TOS Change Detected

' +\n '

Page: ' + p.page_name + '

' +\n '

URL: ' + p.url + '

' +\n '

Detected: ' + p.checked_at + '

' +\n '

AI Summary of Changes

' +\n '

' + summary.replace(/\\n/g, '
') + '

' +\n '

Generated by your TOS Change Watcher workflow in n8n.

';\n\n results.push({\n json: {\n email_subject: 'TOS Change Alert: ' + p.page_name,\n email_body: body,\n url: p.url,\n page_name: p.page_name,\n summary: summary,\n checked_at: p.checked_at,\n new_content: p.new_content\n }\n });\n}\n\nreturn results;" }, "position": [ 1520, 272 ] }, { "id": "d1a1b1c1-0009-4000-8000-000000000009", "name": "Send Change Alert", "type": "integration", "params": { "sendTo": "user@example.com", "message": "={{ $json.email_body }}", "options": { "appendAttribution": false }, "service": "gmail", "subject": "={{ $json.email_subject }}", "operation": "read" }, "position": [ 1744, 272 ] }, { "id": "d1a1b1c1-0010-4000-8000-000000000010", "name": "Log Change", "type": "integration", "params": { "columns": { "value": { "url": "={{ $('Build Email Body').item.json.url }}", "date": "={{ $('Build Email Body').item.json.checked_at }}", "summary": "={{ $('Build Email Body').item.json.summary }}", "page_name": "={{ $('Build Email Body').item.json.page_name }}" }, "schema": [ { "id": "date", "type": "string", "display": true, "required": false, "displayName": "date", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "page_name", "type": "string", "display": true, "required": false, "displayName": "page_name", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "url", "type": "string", "display": true, "required": false, "displayName": "url", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "summary", "type": "string", "display": true, "required": false, "displayName": "summary", "defaultMatch": false, "canBeUsedToMatch": true } ], "mappingMode": "defineBelow" }, "options": {}, "service": "googleSheets", "operation": "append", "sheetName": { "__rl": true, "mode": "name", "value": "Change Log" }, "documentId": { "__rl": true, "mode": "id", "value": "YOUR_GOOGLE_SHEET_ID" } }, "position": [ 1968, 272 ] }, { "id": "d1a1b1c1-0011-4000-8000-000000000011", "name": "Update Page Record", "type": "integration", "params": { "columns": { "value": { "url": "={{ $('Build Email Body').item.json.url }}", "last_checked": "={{ $('Build Email Body').item.json.checked_at }}", "last_content": "={{ $('Build Email Body').item.json.new_content }}" }, "schema": [ { "id": "url", "type": "string", "display": true, "required": false, "displayName": "url", "defaultMatch": true, "canBeUsedToMatch": true }, { "id": "page_name", "type": "string", "display": true, "removed": true, "required": false, "displayName": "page_name", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "last_content", "type": "string", "display": true, "required": false, "displayName": "last_content", "defaultMatch": false, "canBeUsedToMatch": false }, { "id": "last_checked", "type": "string", "display": true, "required": false, "displayName": "last_checked", "defaultMatch": false, "canBeUsedToMatch": false } ], "mappingMode": "defineBelow", "matchingColumns": [ "url" ] }, "options": {}, "service": "googleSheets", "operation": "appendOrUpdate", "sheetName": { "__rl": true, "mode": "name", "value": "Pages" }, "documentId": { "__rl": true, "mode": "id", "value": "YOUR_GOOGLE_SHEET_ID" } }, "position": [ 2192, 272 ] }, { "id": "d1a1b1c1-0012-4000-8000-000000000012", "name": "Update Last Checked", "type": "integration", "params": { "columns": { "value": { "url": "={{ $json.url }}", "last_checked": "={{ $json.checked_at }}", "last_content": "={{ $json.new_content }}" }, "schema": [ { "id": "url", "type": "string", "display": true, "required": false, "displayName": "url", "defaultMatch": true, "canBeUsedToMatch": true }, { "id": "page_name", "type": "string", "display": true, "removed": true, "required": false, "displayName": "page_name", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "last_content", "type": "string", "display": true, "required": false, "displayName": "last_content", "defaultMatch": false, "canBeUsedToMatch": false }, { "id": "last_checked", "type": "string", "display": true, "required": false, "displayName": "last_checked", "defaultMatch": false, "canBeUsedToMatch": false } ], "mappingMode": "defineBelow", "matchingColumns": [ "url" ] }, "options": {}, "service": "googleSheets", "operation": "appendOrUpdate", "sheetName": { "__rl": true, "mode": "name", "value": "Pages" }, "documentId": { "__rl": true, "mode": "id", "value": "YOUR_GOOGLE_SHEET_ID" } }, "position": [ 1312, 704 ] } ] }

Monitor legal policy changes with Google Sheets, Gmail and GPT-4o