Generate personalized sales leads with Claude AI & Explorium for Gmail outreach

Outbound Agent - AI-Powered Lead Generation with Natural Language Prospecting This n8n workflow transforms natural language queries into targeted B2B prospecting campaigns by combining Explorium's data intelligence with AI-powered research and personalized email generation. Simply describe your ideal customer profile in plain English, and the workflow automatically finds prospects, enriches their data, researches them, and creates personalized email drafts. DEMO Template Demo Credentials Required To use this workflow, set up the following credentials in your n8n environment: Anthropic API - Type: API Key - Used for: AI Agent query interpretation, email research, and email writing - Get your API key at Anthropic Console Explorium API - Type: Generic Header Auth - Header: Authorization - Value: Bearer YOURAPIKEY - Used for: Prospect matching, contact enrichment, professional profiles, and MCP research - Get your API key at Explorium Dashboard Explorium MCP - Type: HTTP Header Auth - Used for: Real-time company and prospect intelligence research - Connect to: https://mcp.explorium.ai/mcp Gmail - Type: OAuth2 - Used for: Creating email drafts - Alternative options: Outlook, Mailchimp, SendGrid, Lemlist Go to Settings → Credentials, create these credentials, and assign them in the respective nodes before running the workflow. --- Workflow Overview Node 1: When chat message received This node creates an interactive chat interface where users can describe their prospecting criteria in natural language. - Type: Chat Trigger - Purpose: Accept natural language queries like "Get 5 marketing leaders at fintech startups who joined in the past year and have valid contact information" - Example Prompts: - "Find SaaS executives in New York with 50-200 employees" - "Get marketing directors at healthcare companies" - "Show me VPs at fintech startups with recent funding" Node 2: Chat or Refinement This code node manages the conversation flow, handling both initial user queries and validation error feedback. - Function: Routes either the original chat input or validation error messages to the AI Agent - Dynamic Input: Combines chatInput and errorInput fields - Purpose: Creates a feedback loop for validation error correction Node 3: AI Agent The core intelligence node that interprets natural language and generates structured API calls. Functionality: - Interprets user intent from natural language queries - Maps concepts to Explorium API filters (job levels, departments, company size, revenue, location, etc.) - Generates valid JSON requests with precise filter criteria - Handles off-topic queries with helpful guidance - Connected to MCP Client for real-time filter specifications AI Components: - Anthropic Chat Model: Claude Sonnet 4 for query interpretation - Simple Memory: Maintains conversation context (100 message window) - Output Parser: Structured JSON output with schema validation - MCP Client: Connected to https://mcp.explorium.ai/mcp for Explorium specifications System Instructions: - Expert in converting natural language to Explorium API filters - Can revise previous responses based on validation errors - Strict adherence to allowed filter values and formats - Default settings: mode: "full", size: 10000, pagesize: 100, hasemail: true Node 4: API Call Validation This code node validates the AI-generated API request against Explorium's filter specifications. Validation Checks: - Filter key validity (only allowed filters from approved list) - Value format correctness (enums, ranges, country codes) - No duplicate values in arrays - Proper range structure for experience fields (totalexperiencemonths, currentrolemonths) - Required field presence Allowed Filters: - countrycode, regioncountrycode, companycountrycode, companyregioncountrycode - companysize, companyrevenue, companyage, numberoflocations - googlecategory, naicscategory, linkedincategory, companyname - cityregioncountry, websitekeywords - hasemail, hasphonenumber - joblevel, jobdepart