{ "name": "ComparativeAnalysisAgent", "nodes": [ { "parameters": {}, "type": "n8n-nodes-base.manualTrigger", "typeVersion": 1, "position": [ -1408, -96 ], "id": "ec6646c7-6d97-4576-a26d-3d05fceb8169", "name": "When clicking ‘Execute workflow’" }, { "parameters": { "jsCode": "const subsector = $input.first().json.subsector;\n\n// Peer groups database\nconst peerGroups = {\n \"Cloud Infrastructure\": [\n { rank: 1, ticker: \"MSFT\", company_name: \"Microsoft\" },\n { rank: 2, ticker: \"AMZN\", company_name: \"Amazon\" },\n { rank: 3, ticker: \"GOOGL\", company_name: \"Google\" },\n { rank: 4, ticker: \"ORCL\", company_name: \"Oracle\" }\n ],\n \"Semiconductors\": [\n { rank: 1, ticker: \"NVDA\", company_name: \"NVIDIA\" },\n { rank: 2, ticker: \"AMD\", company_name: \"AMD\" },\n { rank: 3, ticker: \"INTC\", company_name: \"Intel\" },\n { rank: 4, ticker: \"AVGO\", company_name: \"Broadcom\" }\n ],\n \"LLM Developers\": [\n { rank: 1, ticker: \"MSFT\", company_name: \"Microsoft (OpenAI)\" },\n { rank: 2, ticker: \"GOOGL\", company_name: \"Google (Gemini)\" },\n { rank: 3, ticker: \"META\", company_name: \"Meta (Llama)\" },\n { rank: 4, ticker: \"AMZN\", company_name: \"Amazon (Claude/Bedrock)\" }\n ],\n \"Enterprise AI\": [\n { rank: 1, ticker: \"CRM\", company_name: \"Salesforce\" },\n { rank: 2, ticker: \"NOW\", company_name: \"ServiceNow\" },\n { rank: 3, ticker: \"PLTR\", company_name: \"Palantir\" },\n { rank: 4, ticker: \"SNOW\", company_name: \"Snowflake\" }\n ],\n \"AI APIs/Infrastructure\": [\n { rank: 1, ticker: \"MSFT\", company_name: \"Microsoft (Azure AI)\" },\n { rank: 2, ticker: \"GOOGL\", company_name: \"Google (Vertex AI)\" },\n { rank: 3, ticker: \"AMZN\", company_name: \"Amazon (Bedrock)\" },\n { rank: 4, ticker: \"META\", company_name: \"Meta (Llama API)\" }\n ]\n};\n\n// Get companies for selected subsector\nconst companies = peerGroups[subsector];\n\nif (!companies) {\n throw new Error(`Unknown subsector: ${subsector}. Available: ${Object.keys(peerGroups).join(', ')}`);\n}\n\n// Return each company as separate item\nreturn companies.map(company => ({\n json: {\n ticker: company.ticker,\n company_name: company.company_name,\n subsector: subsector,\n rank: company.rank\n }\n}));" }, "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ -992, -96 ], "id": "5f0c3581-76d4-4fdf-8775-34e7963563c9", "name": "Peer Company List" }, { "parameters": { "assignments": { "assignments": [ { "id": "6e74de00-d9d4-4bfe-9750-f8d240ff7253", "name": "subsector", "value": "Cloud Infrastructure", "type": "string" } ] }, "options": {} }, "type": "n8n-nodes-base.set", "typeVersion": 3.4, "position": [ -1200, -96 ], "id": "69f8fb4c-25b7-44f6-a533-2abebf95971c", "name": "Subsector" }, { "parameters": { "jsCode": "// Get all 4 companies from previous node\nconst companies = $input.all().map(item => item.json);\n\n// Get subsector from first company\nconst subsector = companies[0].subsector;\n\n// Store configuration - REPLACE WITH YOUR ACTUAL API KEYS\nconst config = {\n // Current analysis info\n subsector: subsector,\n analysis_start_time: new Date().toISOString(),\n \n // API Keys - IMPORTANT: Replace these with your actual keys\n alpha_vantage_key: '',\n newsapi_key: '',\n groq_api_key: '',\n serp_api_key: '',\n \n // Rate limiting settings\n alpha_vantage_wait_seconds: 13, // Wait 13 seconds between Alpha Vantage calls\n api_call_count: 0,\n \n // Analysis metadata\n total_companies: companies.length\n};\n\n// Add config to each company and return as separate items\nreturn companies.map(company => ({\n json: {\n // Company info\n ticker: company.ticker,\n company_name: company.company_name,\n subsector: company.subsector,\n rank: company.rank,\n \n // Add all config to each company\n ...config\n }\n}));" }, "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ -800, -96 ], "id": "3f92ff52-dd76-493a-865f-cce4d76c126f", "name": "Store Variables & Keys" }, { "parameters": { "url": "=https://www.alphavantage.co/query?function=OVERVIEW&symbol={{ $json.ticker }}&apikey=REPLACE_WITH_CREDENTIAL $json.alpha_vantage_key }}", "options": { "timeout": 30000 } }, "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.2, "position": [ -304, -96 ], "id": "2f6a8f16-287e-4d72-9525-b547bb0f6c9b", "name": "Fetch Financial Overview" }, { "parameters": { "url": "=https://www.alphavantage.co/query?function=TIME_SERIES_DAILY&symbol={{ $json.ticker }}&apikey=REPLACE_WITH_CREDENTIAL $json.alpha_vantage_key }}&outputsize=compact", "options": { "timeout": 30000 } }, "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.2, "position": [ -304, 96 ], "id": "f8e2a86f-5edf-4ae3-9c1d-c31832c76909", "name": "Fetch Financial Time Series" }, { "parameters": { "mode": "combine", "combineBy": "combineByPosition", "options": {} }, "type": "n8n-nodes-base.merge", "typeVersion": 3.2, "position": [ -48, -16 ], "id": "ef75d323-c3c7-4ce6-bceb-4c6941607e0c", "name": "Merge" }, { "parameters": { "jsCode": "// Extract and calculate financial metrics\nconst items = $input.all();\n\nconst results = items.map(item => {\n const data = item.json;\n \n // Helper function to parse numbers safely\n function parseNum(value) {\n if (!value || value === 'None' || value === '-') return null;\n const num = parseFloat(value);\n return isNaN(num) ? null : num;\n }\n \n // Extract ticker (should be in data already)\n const ticker = data.ticker || data.Symbol;\n const companyName = data.company_name || data.Name;\n \n // Extract financial metrics from Overview\n const marketCap = parseNum(data.MarketCapitalization);\n const peRatio = parseNum(data.PERatio);\n const profitMargin = parseNum(data.ProfitMargin);\n const operatingMargin = parseNum(data.OperatingMarginTTM);\n const revenueGrowth = parseNum(data.QuarterlyRevenueGrowthYOY);\n const roe = parseNum(data.ReturnOnEquityTTM);\n const roa = parseNum(data.ReturnOnAssetsTTM);\n const beta = parseNum(data.Beta);\n \n // Calculate price momentum from Daily Prices\n const timeSeries = data['Time Series (Daily)'];\n let priceMomentum30d = null;\n let priceCurrent = null;\n \n if (timeSeries) {\n const dates = Object.keys(timeSeries).sort().reverse();\n priceCurrent = dates[0] ? parseFloat(timeSeries[dates[0]]['4. close']) : null;\n const price30dAgo = dates[29] ? parseFloat(timeSeries[dates[29]]['4. close']) : priceCurrent;\n \n if (priceCurrent && price30dAgo) {\n priceMomentum30d = (priceCurrent - price30dAgo) / price30dAgo;\n }\n }\n \n // Calculate financial health score (0-100)\n // Based on: margins (40%), returns (40%), growth (20%)\n let financialScore = 50; // default\n \n if (profitMargin !== null && roe !== null && revenueGrowth !== null) {\n const marginScore = Math.min(profitMargin * 100, 100) * 0.4;\n const returnScore = Math.min(roe * 100, 100) * 0.4;\n const growthScore = Math.min(Math.max(revenueGrowth * 100, 0), 100) * 0.2;\n \n financialScore = marginScore + returnScore + growthScore;\n }\n \n return {\n json: {\n ticker: ticker,\n company_name: companyName,\n category: 'financial',\n \n // Raw metrics\n market_cap: marketCap,\n pe_ratio: peRatio,\n profit_margin: profitMargin,\n operating_margin: operatingMargin,\n revenue_growth: revenueGrowth,\n roe: roe,\n roa: roa,\n beta: beta,\n \n // Price data\n price_current: priceCurrent,\n price_momentum_30d: priceMomentum30d,\n \n // Calculated score\n financial_score: parseFloat(financialScore.toFixed(1))\n }\n };\n});\n\nreturn results;" }, "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ 160, -16 ], "id": "7d243a4a-61cb-4fef-9f8b-60b28b7f6faa", "name": "Code in JavaScript" }, { "parameters": { "options": {} }, "type": "n8n-nodes-base.splitInBatches", "typeVersion": 3, "position": [ -368, 672 ], "id": "f801b2e2-df7f-4a8c-b94f-c2feb9121309", "name": "Loop - News" }, { "parameters": { "url": "=https://newsapi.org/v2/everything?q=\"{{ $json.company_name }}\"&from={{ $now.minus({days: 28}).toFormat('yyyy-MM-dd') }}&sortBy=relevancy&pageSize=30&language=en&apiKey=REPLACE_WITH_CREDENTIAL $json.newsapi_key }}", "options": { "timeout": 30000 } }, "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.2, "position": [ -112, 912 ], "id": "b336fdad-314a-4656-8fb5-19c30f943015", "name": "HTTP Request" }, { "parameters": { "language": "python", "pythonCode": "# Calculate sentiment and extract company name from query\nitem = _input.first()\ndata = item.json\n\n# Try to get ticker from input\nticker = data.get('ticker', 'UNKNOWN')\n\n# Extract company name from the NewsAPI search metadata\ncompany_name = 'UNKNOWN'\n\n# Method 1: Get from search_parameters if available\nif 'search_parameters' in data:\n query = data['search_parameters'].get('q', '')\n company_name = query.replace('\"', '').strip()\n\n# Method 2: Pattern match from articles if Method 1 fails\nif company_name == 'UNKNOWN' or not company_name:\n # Company name patterns\n company_patterns = {\n 'MSFT': 'Microsoft',\n 'AMZN': 'Amazon',\n 'GOOGL': 'Google',\n 'ORCL': 'Oracle',\n 'NVDA': 'NVIDIA',\n 'AMD': 'AMD',\n 'INTC': 'Intel',\n 'AVGO': 'Broadcom',\n 'META': 'Meta',\n 'CRM': 'Salesforce',\n 'NOW': 'ServiceNow',\n 'PLTR': 'Palantir',\n 'SNOW': 'Snowflake'\n }\n \n # Check first article for company mention\n articles = data.get('articles', [])\n if len(articles) > 0:\n first_article = (articles[0].get('title') or '').lower()\n \n for tick, name in company_patterns.items():\n if name.lower() in first_article:\n ticker = tick\n company_name = name\n break\n\n# Get articles\narticles = data.get('articles', [])\n\n# Sentiment analysis\npositive_words = [\n 'growth', 'beat', 'strong', 'record', 'high', 'gain',\n 'surge', 'innovation', 'profit', 'success', 'breakthrough',\n 'expansion', 'partnership', 'achievement', 'outperform',\n 'upgrade', 'bullish', 'rally', 'soar', 'exceed'\n]\n\nnegative_words = [\n 'decline', 'miss', 'weak', 'fall', 'low', 'risk',\n 'loss', 'concern', 'warning', 'drop', 'lawsuit',\n 'investigation', 'delay', 'problem', 'struggle',\n 'downgrade', 'bearish', 'plunge', 'crash', 'fail'\n]\n\npositive_count = 0\nnegative_count = 0\nneutral_count = 0\nsentiment_sum = 0\n\nfor article in articles:\n title = (article.get('title') or '').lower()\n description = (article.get('description') or '').lower()\n text = title + ' ' + description\n \n pos_score = sum(1 for word in positive_words if word in text)\n neg_score = sum(1 for word in negative_words if word in text)\n \n if pos_score > neg_score:\n positive_count += 1\n sentiment_sum += 1\n elif neg_score > pos_score:\n negative_count += 1\n sentiment_sum -= 1\n else:\n neutral_count += 1\n\n# Calculate scores\navg_sentiment = sentiment_sum / len(articles) if len(articles) > 0 else 0\nsentiment_score = ((avg_sentiment + 1) / 2) * 100\npositive_ratio = positive_count / len(articles) if len(articles) > 0 else 0\n\nreturn [{\n 'json': {\n 'ticker': ticker,\n 'company_name': company_name,\n 'category': 'sentiment',\n \n 'news_articles_30d': len(articles),\n 'positive_articles': positive_count,\n 'negative_articles': negative_count,\n 'neutral_articles': neutral_count,\n 'positive_ratio': round(positive_ratio, 2),\n \n 'sentiment_score': round(sentiment_score, 1),\n 'data_source': 'newsapi'\n }\n}]" }, "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ 112, 912 ], "id": "4b0a7914-6976-4388-9c8a-bc31e53daf8f", "name": "Code in Python (Beta)1" }, { "parameters": { "options": {} }, "type": "n8n-nodes-base.splitInBatches", "typeVersion": 3, "position": [ -272, 384 ], "id": "bf5e3742-103a-4da7-8e8b-362215a87596", "name": "Loop Over Items" }, { "parameters": { "url": "=https://serpapi.com/search.json?engine=google_patents&q={{ $json.company_name }}+artificial+intelligence&num=50&api_key=REPLACE_WITH_CREDENTIAL $json.serp_api_key }}", "options": {} }, "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.2, "position": [ -32, 448 ], "id": "bd6da91b-8e6a-41a5-ac90-241592cd96f6", "name": "HTTP Request1" }, { "parameters": { "language": "python", "pythonCode": "# Process patent data and extract company name from results\nitem = _input.first()\ndata = item.json\n\n# Try to get ticker from input (may or may not exist)\nticker = data.get('ticker', 'UNKNOWN')\n\n# Extract company name from patent assignees\ncompany_name = 'UNKNOWN'\npatents = data.get('organic_results', [])\n\n# Look through patents to find the assignee (company name)\nif len(patents) > 0:\n # Check first few patents for assignee field\n for patent in patents[:5]:\n assignee = patent.get('assignee') or patent.get('assignee_name')\n if assignee:\n company_name = assignee\n break\n \n # If no assignee field, try to extract from snippet/title\n if company_name == 'UNKNOWN':\n # Common company names to look for\n company_patterns = {\n 'MSFT': ['Microsoft', 'Microsoft Corporation'],\n 'AMZN': ['Amazon', 'Amazon.com', 'Amazon Technologies'],\n 'GOOGL': ['Google', 'Google LLC', 'Alphabet'],\n 'ORCL': ['Oracle', 'Oracle Corporation'],\n 'NVDA': ['NVIDIA', 'NVIDIA Corporation'],\n 'AMD': ['Advanced Micro Devices', 'AMD'],\n 'INTC': ['Intel', 'Intel Corporation'],\n 'AVGO': ['Broadcom', 'Broadcom Inc'],\n 'META': ['Meta', 'Facebook', 'Meta Platforms'],\n 'CRM': ['Salesforce', 'Salesforce.com'],\n 'NOW': ['ServiceNow', 'ServiceNow Inc'],\n 'PLTR': ['Palantir', 'Palantir Technologies'],\n 'SNOW': ['Snowflake', 'Snowflake Inc']\n }\n \n # Check first patent's snippet for company name\n first_patent_text = (patents[0].get('snippet') or '').lower()\n \n for tick, names in company_patterns.items():\n if any(name.lower() in first_patent_text for name in names):\n ticker = tick\n company_name = names[0]\n break\n\n# Count total patents\ntotal_patents = len(patents)\n\n# Count recent patents (last 12 months)\nfrom datetime import datetime, timedelta\ntwelve_months_ago = datetime.now() - timedelta(days=365)\nrecent_patents = 0\n\nfor patent in patents:\n date_str = patent.get('filing_date') or patent.get('publication_date')\n if date_str:\n try:\n patent_date = datetime.strptime(date_str.split('T')[0], '%Y-%m-%d')\n if patent_date > twelve_months_ago:\n recent_patents += 1\n except:\n pass\n\ninnovation_velocity = round(recent_patents / 12, 2) if recent_patents > 0 else 0\n\n# Categorize patents\ncategories = {\n 'machine_learning': 0,\n 'computer_vision': 0,\n 'nlp': 0,\n 'robotics': 0,\n 'generative_ai': 0\n}\n\nkeywords = {\n 'machine_learning': ['machine learning', 'neural network', 'deep learning', 'training'],\n 'computer_vision': ['computer vision', 'image', 'visual', 'object detection'],\n 'nlp': ['natural language', 'nlp', 'text', 'language model'],\n 'robotics': ['robot', 'autonomous', 'control'],\n 'generative_ai': ['generative', 'gan', 'diffusion', 'synthesis']\n}\n\nfor patent in patents:\n title = (patent.get('title') or '').lower()\n snippet = (patent.get('snippet') or '').lower()\n text = title + ' ' + snippet\n \n for category, kw_list in keywords.items():\n if any(kw in text for kw in kw_list):\n categories[category] += 1\n\ncategory_diversity = sum(1 for v in categories.values() if v > 0)\n\n# Calculate innovation score (0-100)\nvelocity_score = min((innovation_velocity / 10) * 100, 100) if innovation_velocity > 0 else 0\nvolume_score = min((total_patents / 50) * 100, 100)\ndiversity_score = (category_diversity / 5) * 100\n\ninnovation_score = (\n velocity_score * 0.5 +\n volume_score * 0.3 +\n diversity_score * 0.2\n)\n\nreturn [{\n 'json': {\n 'ticker': ticker,\n 'company_name': company_name,\n 'category': 'patents',\n \n 'total_ai_patents': total_patents,\n 'recent_patents_12m': recent_patents,\n 'innovation_velocity': innovation_velocity,\n 'category_diversity': category_diversity,\n \n 'patents_ml': categories['machine_learning'],\n 'patents_cv': categories['computer_vision'],\n 'patents_nlp': categories['nlp'],\n 'patents_robotics': categories['robotics'],\n 'patents_gen_ai': categories['generative_ai'],\n \n 'innovation_score': round(innovation_score, 1),\n 'data_source': 'google_patents_serpapi'\n }\n}]" }, "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ 192, 448 ], "id": "827be15e-0eeb-48e0-a44b-c51973e30feb", "name": "Code in Python (Beta)" }, { "parameters": { "aggregate": "aggregateAllItemData", "options": {} }, "type": "n8n-nodes-base.aggregate", "typeVersion": 1, "position": [ 192, 256 ], "id": "84691848-a1d2-4bbd-b40d-cae558b808db", "name": "Aggregate" }, { "parameters": { "aggregate": "aggregateAllItemData", "options": {} }, "type": "n8n-nodes-base.aggregate", "typeVersion": 1, "position": [ -32, 704 ], "id": "0dffa3cf-928b-47d5-bb89-55ea69dd76d7", "name": "Aggregate1" }, { "id": "f2d2c400-c082-5ff8-a1d4-7d1eefe3adec", "name": "Publication · Overview and setup", "type": "n8n-nodes-base.stickyNote", "typeVersion": 1, "position": [ -2208, -96 ], "parameters": { "content": "# ComparativeAnalysisAgent\n\n## Purpose and structure\nThis folder contains the **ComparativeAnalysisAgent** workflow. Its executable nodes use `aggregate`, `code`, `httpRequest`, `manualTrigger`, `merge`, `set`, `splitInBatches`. The table and connection map below are derived from the JSON, not an inferred guarantee of behavior.\n\n## Setup\nImport inactive. Configure your own credentials and replace REPLACE_WITH_* values. Review external destinations and Code nodes before running.\n\n## Entry points\nWhen clicking ‘Execute workflow’\n\n## Verification\nSanitized source candidate; not runtime-tested. This overview is generated from node metadata. Read the folder README for the connection map, credential types, and license.\n\n## Attribution\nNik Bear Brown; rights confirmed 2026-10-19\nMIT; preserve source notices.", "width": 700, "height": 800, "color": 1 } } ], "connections": { "When clicking ‘Execute workflow’": { "main": [ [ { "node": "Subsector", "type": "main", "index": 0 } ] ] }, "Subsector": { "main": [ [ { "node": "Peer Company List", "type": "main", "index": 0 } ] ] }, "Peer Company List": { "main": [ [ { "node": "Store Variables & Keys", "type": "main", "index": 0 } ] ] }, "Store Variables & Keys": { "main": [ [ { "node": "Fetch Financial Overview", "type": "main", "index": 0 }, { "node": "Fetch Financial Time Series", "type": "main", "index": 0 }, { "node": "Loop - News", "type": "main", "index": 0 }, { "node": "Loop Over Items", "type": "main", "index": 0 } ] ] }, "Fetch Financial Overview": { "main": [ [ { "node": "Merge", "type": "main", "index": 0 } ] ] }, "Fetch Financial Time Series": { "main": [ [ { "node": "Merge", "type": "main", "index": 1 } ] ] }, "Merge": { "main": [ [ { "node": "Code in JavaScript", "type": "main", "index": 0 } ] ] }, "Loop - News": { "main": [ [ { "node": "Aggregate1", "type": "main", "index": 0 } ], [ { "node": "HTTP Request", "type": "main", "index": 0 } ] ] }, "HTTP Request": { "main": [ [ { "node": "Code in Python (Beta)1", "type": "main", "index": 0 } ] ] }, "Code in Python (Beta)1": { "main": [ [ { "node": "Loop - News", "type": "main", "index": 0 } ] ] }, "Loop Over Items": { "main": [ [ { "node": "Aggregate", "type": "main", "index": 0 } ], [ { "node": "HTTP Request1", "type": "main", "index": 0 } ] ] }, "HTTP Request1": { "main": [ [ { "node": "Code in Python (Beta)", "type": "main", "index": 0 } ] ] }, "Code in Python (Beta)": { "main": [ [ { "node": "Loop Over Items", "type": "main", "index": 0 } ] ] } }, "active": false, "settings": { "executionOrder": "v1" } }