synthesis

Aggregate upstream analysis claims into a cited narrative with consensus notes.

1|Updated Mar 23, 2026
One-click install
npx skills add https://github.com/hellonish/singularity --skill synthesis-hellonish
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: synthesis
Source: https://github.com/hellonish/singularity/tree/main/SKILLS/tier2_analysis/synthesis
Command: npx skills add https://github.com/hellonish/singularity --skill synthesis-hellonish

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Synthesis addresses the need to fuse disparate information streams into a single, coherent narrative. It preserves source citations and avoids cherry-picking, enabling traceable, aggregated insight from multiple upstream analyses.

Core Features & Use Cases

  • Aggregate claims from several upstream nodes (retrieval/analysis) and cluster them by theme.
  • Identify consensus, contradictions, and complementary details, then present a synthesized narrative with explicit citations.
  • Support downstream tasks such as reporting, decision-making, or Q&A pipelines.

Quick Start

Provide at least two upstream analysis outputs containing claims with citations to be synthesized into a unified analysis.

Frequently Asked Questions about synthesis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I synthesize findings from multiple upstream analyses into one narrative?

To synthesize findings from multiple upstream analyses, provide at least two upstream analysis outputs. The Skill aggregates claims, clusters them by theme, and produces a unified narrative with traceable citations and consensus notes.

What is the best way to aggregate evidence and identify consensus across research findings?

The best way to aggregate evidence and identify consensus is by clustering upstream claims by theme. This process highlights contradictions and complementary details, presenting a synthesized narrative with explicit source citations.

How do I combine multiple analysis outputs without cherry-picking data?

You combine multiple analysis outputs without cherry-picking by using a synthesis process that preserves source citations. It fuses disparate information streams into a coherent narrative while maintaining traceable, aggregated insight.

Does evidence synthesis work for risk assessment and decision support scenarios?

Evidence synthesis works for risk assessment and decision-support scenarios. It applies across domains requiring aggregated evidence, turning diverse findings into a cohesive narrative with hedging and clearly linked claims.

How do I maintain citation traceability when merging claims from different sources?

To maintain citation traceability when merging claims, the synthesis process produces an AnalysisOutput with explicitly linked claims and source citations. This avoids cherry-picking and ensures aggregated insight remains traceable.