Help us improve
Share bugs, ideas, or general feedback.
From stepwise-research
Generates standardized Markdown research reports with YAML frontmatter, executive summaries, findings, conclusions, and bibliographies from multi-agent research data.
npx claudepluginhub nikeyes/stepwise-dev --plugin stepwise-researchHow this skill is triggered — by the user, by Claude, or both
Slash command
/stepwise-research:research-reportsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill provides utilities for formatting and managing research reports generated by the stepwise-research plugin.
Generates structured research reports with evidence tables, citations, multi-pass synthesis, and format enforcement for literature reviews, market analysis, or policy briefs.
Orchestrates multi-agent research (Director, Manager, Scouts, Researchers) to produce thorough reports with folder-based output. Useful for deep-dive investigations on any topic.
Summarize deep research results into a markdown report, covering all fields, skipping uncertain values.
Share bugs, ideas, or general feedback.
This skill provides utilities for formatting and managing research reports generated by the stepwise-research plugin.
Claude Code will automatically invoke this skill when:
Purpose: Generate a properly formatted research report with YAML frontmatter, citations, and structured sections.
Usage:
research/skills/research-reports/scripts/generate-report \
--title "Research on [Topic]" \
--query "Original research question" \
--keywords "keyword1,keyword2,keyword3" \
--agent-count N \
--source-count M \
--output-file "thoughts/shared/research/filename.md" \
--executive-summary "Summary text" \
--findings "Findings text with citations" \
--conclusions "Conclusion text" \
--bibliography "Bibliography entries"
Parameters:
--title (required): Report title--query (required): Original research question--keywords (required): Comma-separated keywords--agent-count (required): Number of research agents spawned--source-count (required): Total unique sources cited--output-file (required): Output path (should be in thoughts/shared/research/)--executive-summary (optional): Executive summary section content--findings (optional): Detailed findings section content--conclusions (optional): Conclusions section content--bibliography (optional): Bibliography entries (numbered list)Output: Generates a markdown file with this structure:
---
title: [Title]
date: YYYY-MM-DD
query: [Query]
keywords: [keywords]
status: complete
agent_count: N
source_count: M
---
# [Title]
## Executive Summary
[Content]
## Detailed Findings
[Content with citations]
## Conclusions
[Content]
## Bibliography
[Numbered citations]
---
*Research conducted by stepwise-research multi-agent system*
*Generated: [timestamp]*
title: Human-readable report titledate: ISO 8601 date (YYYY-MM-DD)query: Original research question (verbatim)keywords: 5-8 extracted key termsstatus: complete | draft | in-progressagent_count: Number of research agents usedsource_count: Total unique sources cited[N] Source Title - URLCitations must follow this format:
In-text:
Docker uses containerd as its default runtime [1] [2].
Bibliography:
[1] Docker Documentation - https://docs.docker.com/engine/
[2] Containerd Official Site - https://containerd.io/
Reports are saved to:
thoughts/shared/research/[sanitized-topic]-[YYYY-MM-DD].md
Filename sanitization rules:
Example:
what-is-kubernetes-and-how-does-it-work-2026-02-19.mdAfter report creation, the thoughts-management Skill will automatically sync hardlinks to thoughts/searchable/.
The generate-report script:
If report generation fails:
thoughts/shared/research/ directory existsPotential future additions to this skill:
validate-report: Check report structure and citation formatexport-report: Convert to PDF, HTML, or other formatsmerge-reports: Combine multiple research reportsextract-citations: Pull bibliography from existing reports