Architect Comprehensive Researcher

Multi-source investigative research agent that decomposes complex topics into structured research questions, evaluates source credibility, and delivers publication-ready reports for podcast content planning.

When to Use This Agent

Choose this agent when you need to:

• Conduct broad investigative research spanning academic, industry, government, and media sources
• Produce structured research reports with executive summaries, inline citations, and confidence assessments
• Evaluate controversial or multi-perspective topics with balanced presentation of competing viewpoints
• Build comprehensive background dossiers for podcast episode planning across any domain

Consider alternatives when:

• You need narrowly focused academic literature review with arXiv and Semantic Scholar depth (use the Academic Research Synthesizer Copilot)
• You are coordinating episode production logistics rather than content research (use the Episode Orchestrator Pro)

Quick Start

Configuration

Copy
name: architect-comprehensive-researcher
type: agent
category: podcast-creator-team

Example Invocation

Copy
claude agent:invoke architect-comprehensive-researcher "Research the global state of nuclear fusion energy for a 3-episode podcast series"

Example Output

Comprehensive Research Report — Nuclear Fusion Energy (2024-2026)
Research Questions: 7 | Sources Consulted: 42 | Report Sections: 5

Executive Summary:
- Private fusion companies raised $6.2B in 2025, up 40% YoY
- NIF achieved net energy gain in Dec 2022; no commercial reactor expected before 2035
- Regulatory frameworks remain fragmented across US, EU, and IAEA jurisdictions
- Public sentiment broadly positive but skepticism about timeline claims persists

Confidence Assessment:
  Strong evidence: Investment figures, NIF milestone
  Moderate evidence: Timeline projections (conflicting expert opinions)
  Weak evidence: Cost-per-MWh estimates (insufficient operational data)

Source types: 12 academic papers, 8 govt reports, 14 industry analyses, 8 journalism pieces
Full bibliography: 42 entries (Chicago format)

Core Concepts

Research Methodology Overview

Aspect	Details
Question Generation	5-8 specific sub-questions covering history, current state, key players, controversies, and outlook
Source Diversity	Academic, government, industry, journalism, and primary sources for triangulation
Credibility Evaluation	Assess each source for bias, methodology, recency, author expertise, and funding conflicts
Report Structure	Executive summary, thematic body sections, limitation acknowledgment, and full bibliography

Investigative Research Architecture

┌─────────────────┐     ┌─────────────────┐
│  Topic Input     │────▶│  Research       │
│  from Producer   │     │  Question Gen   │
└─────────────────┘     └─────────────────┘
        │                       │
        ▼                       ▼
┌─────────────────┐     ┌─────────────────┐
│  Multi-Source    │────▶│  Credibility    │
│  Parallel Search │     │  Filter & Rank  │
└─────────────────┘     └─────────────────┘
        │                       │
        ▼                       ▼
┌─────────────────┐     ┌─────────────────┐
│  Cross-Source    │────▶│  Structured     │
│  Triangulation   │     │  Report Builder │
└─────────────────┘     └─────────────────┘

Configuration

Parameter	Type	Default	Description
research_questions	integer	7	Number of sub-questions to generate per topic for comprehensive coverage
min_sources_per_question	integer	3	Minimum distinct sources to consult per research question
source_diversity	boolean	true	Enforce that sources span at least 3 different types (academic, govt, industry, etc.)
citation_style	string	Chicago	Citation format for inline references and bibliography
conflict_of_interest_check	boolean	true	Flag sources with potential funding or organizational bias

Best Practices

1. Decompose Before Searching Resist the urge to search the broad topic immediately. Spending time formulating precise, answerable sub-questions dramatically improves search result relevance and ensures comprehensive coverage. A well-decomposed topic naturally produces a well-structured report because each question maps to a report section.

2. Triangulate Claims Across Source Types A claim supported by an academic paper, a government report, and an industry analysis is far more credible than one supported by three papers from the same research group. Source-type diversity provides structural protection against systematic bias and strengthens the report's authority.

3. Present Disagreements as Features, Not Bugs When experts disagree, the disagreement itself is valuable podcast content. Present competing viewpoints with the reasoning behind each position and the evidence each side cites. This honest framing gives podcast hosts the material to facilitate nuanced discussions rather than presenting false certainty.

4. Quantify Evidence Strength Explicitly Use phrases like "strong evidence from 8 independent studies," "preliminary findings from a single pilot program," or "expert consensus without empirical validation." These qualifiers help podcast hosts calibrate how definitively they can state claims during recording and protect the show's credibility.

5. Document Gaps and Suggest Follow-Up Research Every comprehensive report will encounter questions that available sources cannot fully answer. Documenting these gaps explicitly serves two purposes: it prevents the podcast from overclaiming, and it identifies potential expert interview questions that could fill the gaps during guest segments.

Common Issues

1. Source bias not detected when all sources share the same funding origin Industry-funded research may present consistent but systematically biased conclusions. Enable conflict_of_interest_check and examine funding disclosures and organizational affiliations. When all favorable evidence traces to the same funding source, downgrade confidence and note the pattern in the report.

2. Report becomes excessively long and loses editorial focus Comprehensive does not mean exhaustive. When a report exceeds 15 pages, review each section for information that is interesting but not essential to the podcast topic. Move tangential details to an appendix and keep the main report focused on the 5-7 most important findings that drive the episode narrative.

3. Recency bias overshadowing foundational context Focusing exclusively on recent sources can leave out foundational work that provides essential context. For any topic, include 2-3 seminal references that established the field's core concepts, even if they are older, so the podcast can ground its discussion in established knowledge before exploring recent developments.