data-synthesis

Data Synthesis Skill

This skill provides a structured approach to combining research findings from multiple sources into unified, coherent insights. It ensures that synthesized outputs maintain accuracy, proper attribution, and analytical depth.

When to Use This Skill

Activate this skill in the following scenarios:

Primary Use Cases

• Multi-source integration: Combining findings from 2+ research sub-agents working on related topics
• Cross-thread synthesis: Merging parallel research threads into a unified narrative
• Comparative analysis: Analyzing similarities and differences across multiple data sources
• Executive summaries: Condensing extensive research into actionable insights
• Gap identification: Finding missing information by comparing what different sources cover

Trigger Conditions

• You have gathered research from multiple sub-agents or tools
• Sources contain overlapping but not identical information
• The user needs a cohesive understanding rather than fragmented data points
• Contradictory information needs to be reconciled
• Multiple perspectives on a topic need to be unified

Synthesis Workflow

Follow this step-by-step process for effective data synthesis:

Step 1: Gather All Source Materials

Before beginning synthesis, ensure you have:

1. Inventory all sources

   • List every research output, file, or response to be synthesized
   • Note the origin (sub-agent ID, tool used, timestamp)
   • Record the scope and focus of each source

2. Assess source quality

   • Evaluate reliability (primary vs secondary sources)
   • Check recency and relevance
   • Identify potential biases or limitations

3. Organize by category

   /research_workspace/synthesis/
   ├── sources/
   │   ├── agent_1_findings.md
   │   ├── agent_2_findings.md
   │   └── agent_3_findings.md
   ├── working/
   │   └── theme_analysis.md
   └── output/
       └── synthesized_report.md
   

Step 2: Identify Common Themes and Contradictions