WR
write-findings
Research Findings Documentation and Analysis Reporting. Use when user needs to document experimental findings, analysis results, and insights in clear, evidence-based reports.
Install
mkdir -p .claude/skills/write-findings && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/15776" && unzip -o skill.zip -d .claude/skills/write-findings && rm skill.zipInstalls to .claude/skills/write-findings
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Research Findings Documentation and Analysis Reporting. Use when user needs to document experimental findings, analysis results, and insights in clear, evidence-based reports.175 chars✓ has a “when” trigger
About this skill
Workflow: Findings Writer
You are an expert at documenting experimental findings, analysis results, and insights in clear, evidence-based reports.
Goal
Create comprehensive findings documents that clearly communicate discoveries, validate claims with evidence, and provide actionable insights.
Report Structure
- Executive Summary: Key findings in 2-3 sentences
- Background: Context and motivation
- Methodology: Approach used (reference implementation plan)
- Results: Data-driven findings with tables/figures
- Analysis: Interpretation of results
- Limitations: Honest assessment of constraints
- Conclusions: Summary and implications
- Recommendations: Next steps or applications
Writing Standards
- Use clear, professional language
- Link every numeric claim to an artifact (CSV, JSON, log)
- Include confidence intervals and uncertainty estimates
- Use tables and figures to support claims
- Acknowledge limitations and negative results
- Define any specialized terminology
Content Checklist
- All claims have evidence (artifact reference)
- Numerical results include uncertainty/confidence intervals
- Tables and figures are clear and properly labeled
- Methodology is reproducible
- Limitations are explicitly stated
- Negative results are reported honestly
- Conclusions don't overstate evidence
- Recommendations are actionable
Example Structure
# Findings Report: [Experiment Name]
## Executive Summary
[2-3 sentence summary of key discoveries]
## Background
[Why this experiment matters, what problem it addresses]
## Methodology
Reference: `docs/implementation_plan/00_overview.md`
[Brief overview, link to detailed plan]
## Results
[Tables and figures showing key metrics]
See: `results/tables/final_summary_tables.csv`
### Key Finding 1
[Description with supporting data and confidence intervals]
### Key Finding 2
[Description with supporting data]
## Analysis
[Interpretation of results, why patterns emerged]
## Limitations
[Constraints, threats to validity, caveats]
## Conclusions
[Summary of key takeaways]
## Recommendations
[What to do with these findings]
Evidence References
Always cite artifacts:
results/tables/*.csv- Tabular resultsresults/metrics/*.json- Metric detailsresults/figures/*.png- Visualizationsresults/logs/*.log- Execution logsdocs/implementation_plan/**- Methodology