What does this skill do?

Operational Research Skill acts as an orchestration layer on top of the research stack, determining when and how to combine tools such as exa-search, deep-research, or market-research to obtain fast, reliable, and traceable answers. It classifies each request, selects the most efficient evidence path, and generates reports that clearly distinguish between facts with sources, inferences, and actionable recommendations.

Supplier Comparison
Compare platforms or products and generate a ranking with actionable recommendations using exa-search, deep-research, and market-research.
Contact Enrichment
It standardizes lists of companies or contacts provided by the user and ranks them by interest score using lead intelligence.
Quick, Fact-Based Answers
Get up-to-date, verifiable data that isn't limited to the local context, with specific dates for each time-sensitive statement.
Recurring Monitoring
It identifies recurring queries and recommends converting them into automated workflows using knowledge-ops instead of repeating manual searches.

Usage examples

⚖️ Provider Comparison
Compare the three leading observability platforms for a company with 50 engineers.
👥 Contact Enrichment
Here's a list of 20 digital health startups. Let me know which ones are worth reaching out to.
🔎 Latest Data
What's new in AI regulation in the EU? Send me a memo with sources and dates.
🤝 Contact Recommendations
Who should I talk to in order to understand the cybersecurity market in Spain?

Features

Multi-tool orchestration It combines exa-search, deep-research, market-research, and lead-intelligence based on the type of query and scales only when necessary.
Automatic Classification of Requests Choose the appropriate track before searching: quick facts, comparison memo, contact enrichment, or monitoring candidate.
Explicit boundaries of evidence Separate facts—including sources and dates, user-provided context, model inferences, and recommendations—into distinct blocks.
A More Streamlined Evidence Trail Start with exa-search for quick discovery, and scale up to deep-research or market-research only when multi-source synthesis calls for it.
Identifying Recurring Inquiries Identify recurring patterns and recommend automating them using knowledge-ops to store the results in a long-term context.

Frequently asked questions

exa-search Orchestra for rapid discovery, deep-research for synthesis with citations, market-research for rankings and recommendations, lead-intelligence for contact enrichment, and knowledge-ops for storing lasting context.
Yes. This skill is a decision-making layer that coordinates the tools that should be available in your Claude environment. It does not replace the individual tools; rather, it decides when and how to combine them.
Each response organizes the output into blocks: EVIDENCE (facts with sources and dates), INFERENCE (deductions from the model), and RECOMMENDATION (next steps and monitoring suggestions, if applicable).
Yes. The skill categorizes any material you provide into three groups: facts that have already been verified, facts that require verification, and open-ended questions. It does not restart the analysis from scratch if you have already built part of the model.
Operations Research — Research Orchestration with Claude AI

¿Prefieres escuchar el contenido? Genera la narración de audio con un clic.