agentleFS
rongarederongarede/claude-skills-research/skills/research-analyst/SKILL.md

research-analyst

Skill3 starsChanged 8 months agoMITView on GitHub

Structured research analysis with hypothesis formation, evidence evaluation, and actionable recommendations. Use when users need systematic analysis of a topic, technology evaluation, feasibility studies, literature synthesis, or evidence-based decision support. Complements deep-research by focusing on analytical reasoning rather than report formatting.

10 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 6 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Always produce a structured document with:

    line 119
  2. Very specific:

    Use deep-research for initial evidence collection

    line 131
  3. Very specific:

    Use deep-think for complex reasoning about ambiguous findings

    line 132
  4. Very specific:

    Use tech-scout for technology-specific evaluations

    line 133
  5. Distinctive:

    Do NOT present opinions as findings

    line 138
  6. Distinctive:

    Do NOT cherry-pick evidence to support a predetermined conclusion

    line 139
  7. Distinctive:

    Do NOT ignore contradictory evidence

    line 140
  8. Distinctive:

    Do NOT conflate correlation with causation

    line 141
  9. Distinctive:

    Do NOT make recommendations without stating confidence level

    line 142

“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Research Analyst

Systematic research analysis skill that forms hypotheses, evaluates evidence, and produces structured analytical assessments.

Trigger Words

  • /research-analyst
  • "analyze this topic"
  • "evaluate the evidence"
  • "what does the research say about"
  • "feasibility analysis"
  • "systematic analysis"

Quick Start

  1. Define the research question and analytical scope
  2. Form initial hypotheses or analytical framework
  3. Gather and evaluate evidence (quality, relevance, recency)
  4. Synthesize findings into structured analysis
  5. Produce actionable recommendations with confidence levels

Core Workflow

Research Analyst Progress:
- [ ] Step 1: Define research question and scope
- [ ] Step 2: Establish analytical framework
- [ ] Step 3: Evidence gathering and evaluation
- [ ] Step 4: Cross-reference and validate findings
- [ ] Step 5: Synthesize analysis with confidence ratings
- [ ] Step 6: Produce recommendations

Step 1: Define Research Question

  • Clarify the specific question or problem to analyze
  • Identify constraints: time, domain, depth
  • Determine the decision this analysis will inform
  • Identify key stakeholders and their information needs

Step 2: Analytical Framework

Choose and apply one or more frameworks:

FrameworkBest For
SWOTStrategic positioning, competitive analysis
PESTELMacro-environment, policy impact
Porter's Five ForcesIndustry/market dynamics
Root Cause AnalysisProblem diagnosis
Cost-BenefitInvestment/decision justification
Comparative MatrixTechnology/tool evaluation
Literature SynthesisAcademic/research topics

Step 3: Evidence Gathering

For each piece of evidence, record:

| # | Source | Type | Quality | Relevance | Key Finding |
|---|--------|------|---------|-----------|-------------|
| 1 | [source] | [primary/secondary/tertiary] | [H/M/L] | [H/M/L] | [finding] |

Quality Assessment Criteria:

  • High: Peer-reviewed, primary data, recent (< 2 years), reputable source
  • Medium: Industry reports, established blogs, conference papers
  • Low: Anecdotal, outdated (> 5 years), unverified claims

Step 4: Cross-Reference and Validate

  • Triangulate findings across multiple sources
  • Identify contradictions and resolve them
  • Note gaps in available evidence
  • Flag assumptions that cannot be verified

Step 5: Synthesis

Structure the analysis as:

## Analysis Summary

### Key Findings
1. [Finding with confidence level: HIGH/MEDIUM/LOW]
2. [Finding with confidence level]

### Evidence Strength
- Strong evidence supports: [claims]
- Moderate evidence suggests: [claims]
- Weak/insufficient evidence for: [claims]

### Contradictions and Uncertainties
- [Area of disagreement and possible explanations]

### Limitations
- [Scope limitations, data gaps, methodology constraints]

Step 6: Recommendations

## Recommendations

| Priority | Recommendation | Confidence | Evidence Basis | Risk if Ignored |
|----------|---------------|------------|----------------|-----------------|
| P1 | [action] | HIGH | [refs] | [risk] |
| P2 | [action] | MEDIUM | [refs] | [risk] |

Output Format

Always produce a structured document with:

  1. Executive Summary (2-3 sentences)
  2. Research Question (explicit statement)
  3. Methodology (frameworks used, sources consulted)
  4. Findings (organized by theme or hypothesis)
  5. Analysis (synthesis with confidence levels)
  6. Recommendations (prioritized, actionable)
  7. Appendix (evidence table, source list)

Integration with Other Skills

  • Use deep-research for initial evidence collection
  • Use deep-think for complex reasoning about ambiguous findings
  • Use tech-scout for technology-specific evaluations
  • Feed results to brainstorm for solution ideation

Anti-Patterns

  • Do NOT present opinions as findings
  • Do NOT cherry-pick evidence to support a predetermined conclusion
  • Do NOT ignore contradictory evidence
  • Do NOT conflate correlation with causation
  • Do NOT make recommendations without stating confidence level

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.