Exa Websets MCP Server

# Exa Websets Orchestration Workflows Three powerful orchestration prompts guide you through common use cases for Exa's websets platform. These prompts provide step-by-step workflows with parameterized guidance tailored to your specific needs. ## Quick Access All orchestrations are available as MCP prompts and can be accessed through any compatible MCP client. ### 1. Marketing Orchestration **For**: Competitor monitoring, brand health tracking, share of voice analysis **Prompt Name**: `marketing_orchestration` **Parameters**: - `companyName` (required): Your company name - `targetAudience` (optional): Target audience segment for monitoring - `timeframe` (optional): Time range for monitoring (e.g., '30d', '90d') **Example MCP Call**: ```json { "method": "prompts/get", "params": { "name": "marketing_orchestration", "arguments": { "companyName": "TechCorp", "targetAudience": "Enterprise DevOps", "timeframe": "30d" } } } ``` **What You'll Get**: - Step-by-step competitor monitoring workflow - Competitor discovery queries - Brand health tracking methodology - Share of voice analysis approach - Content gap identification - Ready-to-use tool commands --- ### 2. CRM Orchestration **For**: Lead discovery, account-based marketing, intent scoring **Prompt Name**: `crm_orchestration` **Parameters**: - `idealCustomerProfile` (required): Description of your ideal customer profile (ICP) - `region` (optional): Geographic region for lead search (e.g., 'North America', 'Europe') - `intentSignals` (optional): Comma-separated intent signals (e.g., 'hiring,funding,product launches') **Example MCP Call**: ```json { "method": "prompts/get", "params": { "name": "crm_orchestration", "arguments": { "idealCustomerProfile": "Mid-market SaaS with 50-500 employees", "region": "North America", "intentSignals": "hiring,funding" } } } ``` **What You'll Get**: - Lead discovery workflow with search strategies - ICP matching criteria and scoring - Intent signal identification - Enrichment strategies for lead data - CRM integration guidelines - Ready-to-use search and enrichment commands --- ### 3. Hiring Orchestration **For**: Recruiter candidate sourcing and job seeker target company tracking **Prompt Name**: `hiring_orchestration` **Parameters**: - `mode` (required): Either `"recruiter"` or `"job_seeker"` - `role` (required): Job title or role to search for - `location` (optional): Location preference (e.g., 'Remote', 'Bay Area', 'NYC') - `seniority` (optional): Seniority level (e.g., 'junior', 'mid-level', 'senior') - `keywords` (optional): Comma-separated key skills/interests **Example - Recruiter Mode**: ```json { "method": "prompts/get", "params": { "name": "hiring_orchestration", "arguments": { "mode": "recruiter", "role": "Senior Software Engineer", "location": "Remote", "seniority": "senior", "keywords": "Go,Kubernetes,distributed systems" } } } ``` **Example - Job Seeker Mode**: ```json { "method": "prompts/get", "params": { "name": "hiring_orchestration", "arguments": { "mode": "job_seeker", "role": "Product Manager", "location": "San Francisco", "keywords": "AI/ML,B2B,SaaS" } } } ``` **What You'll Get**: - **Recruiter Mode**: Passive candidate sourcing workflow, search strategies, networking approaches - **Job Seeker Mode**: Target company tracking, company research methodology, networking strategies - Comprehensive search queries - Enrichment approaches - Ready-to-use tool commands --- ## How These Prompts Work Each orchestration prompt: 1. **Explains the workflow** - Clear step-by-step approach tailored to your parameters 2. **Provides search queries** - Ready-to-use JSON examples for websets_manager tool 3. **Includes examples** - Real-world scenarios showing how to use each step 4. **Offers best practices** - Tips for optimal results with your specific use case 5. **Suggests enrichments** - How to enhance and analyze results 6. **Troubleshoots** - Common issues and how to resolve them --- ## Integration with Other Tools These orchestrations work seamlessly with the following MCP tools: - **websets_manager**: Create websets, search within them, and enrich data - **web_search_exa**: Perform real-time web searches for discovery - **knowledge_graph**: Track connections and insights across multiple websets --- ## Example Workflow: Marketing Use Case A marketer wants to monitor a competitor (Acme Corp) in the DevOps space over 30 days: 1. **Call the orchestration**: ```json { "method": "prompts/get", "params": { "name": "marketing_orchestration", "arguments": { "companyName": "Your Company", "targetAudience": "DevOps", "timeframe": "30d" } } } ``` 2. **Follow the workflow steps** provided in the response: - Create a webset for Acme Corp competitor mentions - Search for Acme Corp announcements and press releases - Monitor for product launches - Track mentions in tech communities - Analyze share of voice vs competitors 3. **Use the provided tool calls** to execute each step with websets_manager and web_search_exa 4. **Enrich and analyze** using the suggested enrichment strategies 5. **Iterate** based on findings to refine your research --- ## Support and Troubleshooting ### Common Issues **Issue**: "I'm not getting relevant results" - **Solution**: Review the "Troubleshooting" section in the orchestration for query refinement tips - **Next Step**: Try using different intent signals or search keywords **Issue**: "Results are too broad or too narrow" - **Solution**: Adjust your parameters (e.g., narrow the region, add more specific keywords) - **Next Step**: Create separate websets for different segments **Issue**: "I need more context on a specific step" - **Solution**: Each orchestration includes an "Examples" section with real-world scenarios - **Next Step**: Look for examples matching your specific use case --- ## Best Practices 1. **Start with core parameters** - Use the required parameters first, then add optional ones 2. **Review all examples** - Each orchestration includes 3+ real-world examples 3. **Use the workflow iteratively** - Complete one step, review results, adjust next step 4. **Combine with enrichment** - Use websets_manager enrichment features for deeper insights 5. **Track with knowledge_graph** - Connect findings across multiple websets for better analysis --- ## Future Enhancements We're continuously improving these orchestrations: - [ ] Integration with CRM platforms (Salesforce, HubSpot) - [ ] Advanced filtering and deduplication - [ ] Scheduled webset updates and monitoring - [ ] Automated alerting on key findings - [ ] Custom industry-specific workflows