conducting-deep-research

Produce iterative research reports with source scoring and contradiction handling.

6|2|Updated Dec 19, 2025
One-click install
npx skills add https://github.com/synaptiai/synapti-marketplace --skill conducting-deep-research
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: conducting-deep-research
Source: https://github.com/synaptiai/synapti-marketplace/tree/main/plugins/decipon/skills/deep-research
Command: npx skills add https://github.com/synaptiai/synapti-marketplace --skill conducting-deep-research

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the need for thorough, multi-source research and verifiable synthesis when answering complex questions that require iterative evidence gathering, contradiction handling, and clear confidence judgments. It reduces the risk of unsupported claims and single-source bias by enforcing an iterative draft→critique→targeted-research→refine workflow and structured source scoring.

Core Features & Use Cases

  • Iterative Methodology: Produces an initial noisy draft, applies red-team critique, performs targeted web searches with explicit reflection after each search, then refines until quality converges.
  • Structured Evidence Handling: Scores sources, tracks contradictions, and documents facts with provenance and confidence ratings for transparent attribution.
  • Templates & Quality Controls: Includes research-brief templates, report templates, a quality log with iteration scoring, and stopping/iteration rules for repeatable outputs.
  • Interactive Decision Points: Uses explicit user prompts at scope, brief validation, iteration continuation, and contradiction resolution stages to align research trade-offs.
  • Use Cases: Due diligence, competitive and technology assessments, state-of-the-art surveys, policy or market research, and comprehensive comparative analyses.

Quick Start

Ask the agent to produce a comprehensive, well-sourced research brief and iterative report on the current state of nuclear fusion, including sources, confidence scores, contradictions, and a three-iteration quality log.

Frequently Asked Questions about conducting-deep-research

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

FAQPage Schema
How do I generate a deep research report with source evaluation and contradiction handling?

Deep research reports are produced through an iterative workflow of drafting, red-team critique, targeted web searches, and refinement. The process scores sources, handles contradictions, and applies explicit reflection to ensure verifiable synthesis and clear confidence judgments for complex questions.

What is iterative research synthesis and when do I need it for due diligence?

Iterative research synthesis is a methodology that repeatedly drafts, critiques, and refines information through targeted web searches. You need it for due diligence, comparative analyses, and technology assessments where single-source bias is a risk and verifiable, multi-source evidence gathering is required.

Can I use automated web search to resolve contradictions in comparative analyses?

Yes, you can use automated web search to resolve contradictions by applying targeted queries after an initial draft and red-team critique. The workflow explicitly tracks contradictions, uses interactive decision points for resolution, and applies source scoring to ensure transparent attribution.

What is the best way to structure a state-of-the-art survey with confidence scores?

The best way to structure a state-of-the-art survey is using the Skill's built-in research-brief and report templates. These templates enforce structured evidence handling, documenting facts with provenance, assigning confidence ratings, and including a quality log with iteration scoring for repeatable outputs.

Does deep research support configurable search budgets and stopping rules?

Yes, deep research supports configurable search budgets and explicit stopping or iteration rules. These quality controls ensure repeatable outputs by defining when the iterative draft, critique, and refinement cycles should converge based on quality scoring.