research-scoping — Claude Code skill: what it does, how to install
Previews and maps research questions in one pass before running heavy workflows. Helps spot tensions and decide whether to escalate.
npx skills add gasserane/personal-skills --skill research-scoping -a claude-codeWhat research-scoping does
This skill analyzes complex evaluation or MEL inquiries inside a single prompt pass. Instead of launching heavy agent loops right away, it borrows the multi-angle tactic from Stanford's STORM framework. The model extracts between 4 and 6 distinct viewpoints. It outlines their primary friction points and sketches an outline for an eventual report. It then delivers a SIMPLE or COMPLEX triage verdict.
It runs without subagents. The design deliberately avoids literature searches and empirical claims. Instead of asserting factual dates, partner records, or metrics, the output inserts a ⚠️ to verify: tag. If a project requires evidence and citations, the resulting map gets handed off to tools like /ann or /researcher.
What it helps with
- map conflicting analytical perspectives before launching costly agent workflows
- detect live tensions across 4 to 6 viewpoints on an evaluation topic
- generate a structured report outline without performing web searches
- triage research questions into simple inline answers or complex escalations
Example prompts for the agent
- Map the tensions around our proposed evaluation design before we run full research.
- Scope this learning question across 4 to 6 angles and give me a triage verdict.
- Run a fast scoping pass on this project proposal without looking up citations.
How to install research-scoping
1Skills CLI (Vercel)
npx skills add gasserane/personal-skills --skill research-scoping -a claude-code2Manually: put the files in place
~/.claude/skills/research-scoping/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/research-scoping/
1Skills CLI (Vercel)
npx skills add gasserane/personal-skills --skill research-scoping -a codex2Manually: put the files in place
.agents/skills/research-scoping/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add gasserane/personal-skills --skill research-scoping -a cursor2Manually: put the files in place
.cursor/skills/research-scoping/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install gasserane/personal-skills research-scoping2Manually: put the files in place
.github/skills/research-scoping/SKILL.md1Skills CLI (Vercel)
npx skills add gasserane/personal-skills --skill research-scoping -a gemini-cli2Platform CLI
gemini skills install https://github.com/gasserane/personal-skills --path skills/research-scopingManually: put the files in place
.devin/skills/research-scoping/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add gasserane/personal-skills --skill research-scoping -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-scoping
How to install research-scoping in Claude Code?
Run npx skills add gasserane/personal-skills --skill research-scoping -a claude-code. The file will be placed in ~/.claude/skills/research-scoping/SKILL.md.
What does research-scoping do?
It maps a research question within a single prompt context. It uncovers disagreeing viewpoints, drafts a skeleton structure, and issues an escalation recommendation.
Why does research-scoping avoid citations?
The skill avoids citations so unverified statements never look authoritative. Every output line serves strictly as a hypothesis marked for later verification.
