research-executor — Claude Code skill: 7-phase deep research
The research-executor skill automates multi-stage research: breaks a question into subtopics, deploys parallel agents to gather data from multiple sources, cross-validates findings, and generates a report with citations.
no license 114 installs ★ 379 updated 2026-03-07
npx skills add liangdabiao/Claude-Code-Stock-Deep-Research-Agent --skill research-executor -a claude-codeWhat research-executor does
research-executor implements a 7-phase deep research workflow using Graph of Thoughts. It takes a structured research request, decomposes it into 3–7 subtopics, then launches multiple agents in parallel: web researchers gather current information and trends, technical agents search for papers and specifications, a cross-reference agent validates claims across sources.
Every claim in the output report includes a citation: author, date, source title, URL or DOI. The skill rates sources on an A–E scale, from peer-reviewed RCTs down to anecdotal evidence, and flags contradictions between sources. Output is a folder with executive summary, full report, data files, visuals, source inventory, and appendices.
Use this for complex topics, high-stakes research, or exploratory work where evidence matters. Requires a structured prompt with research questions and constraints.
What it helps with
- search multiple angles of a topic without manually visiting each source
- compile a report where every fact has a citation
- verify facts against independent sources
- rate source quality and surface disagreements
- package findings into an organized folder structure
Example prompts for the agent
- Run a 7-phase market research on AI coding platforms: identify key players, pricing tiers, licensing trends over the past year
- Research OAuth 2.0 security: how it works, common vulnerabilities, testing approaches, and how major companies address it
- Analyze the impact of remote work on development productivity: collect data from reports, academic studies, and case reports
How to install research-executor
1Skills CLI (Vercel)
npx skills add liangdabiao/Claude-Code-Stock-Deep-Research-Agent --skill research-executor -a claude-code2Manually: put the files in place
~/.claude/skills/research-executor/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/research-executor/
1Skills CLI (Vercel)
npx skills add liangdabiao/Claude-Code-Stock-Deep-Research-Agent --skill research-executor -a codex2Manually: put the files in place
.agents/skills/research-executor/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add liangdabiao/Claude-Code-Stock-Deep-Research-Agent --skill research-executor -a cursor2Manually: put the files in place
.cursor/skills/research-executor/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install liangdabiao/Claude-Code-Stock-Deep-Research-Agent 1.claude/skills/research-executor2Manually: put the files in place
.github/skills/research-executor/SKILL.md1Skills CLI (Vercel)
npx skills add liangdabiao/Claude-Code-Stock-Deep-Research-Agent --skill research-executor -a gemini-cli2Platform CLI
gemini skills install https://github.com/liangdabiao/Claude-Code-Stock-Deep-Research-Agent --path 1.claude/skills/research-executorManually: put the files in place
.devin/skills/research-executor/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add liangdabiao/Claude-Code-Stock-Deep-Research-Agent --skill research-executor -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about research-executor
How do I install research-executor in Claude Code?
Run npx skills add liangdabiao/Claude-Code-Stock-Deep-Research-Agent --skill research-executor -a claude-code in your terminal, then open ~/.claude/skills/research-executor/SKILL.md. Other platforms use different commands: Codex uses .agents/skills/, Cursor uses .cursor/skills/, GitHub Copilot uses gh skill install.
What does research-executor do?
It handles breaking down a complex question into research subtasks, running parallel searches across different sources, validating findings against each other, and writing a report with citations. It saves time on manual synthesis when you need evidence backing each claim.
What output will I get?
A folder containing an executive summary, full report, data tables, visuals, source inventory with quality ratings (A–E), and appendices. Each fact includes author, date, and a link to the source.