deep-research · lv416e — Claude Code skill: what it does, how to install
deep-research turns a vague “look into this” request into a staged research run: plan, gather from the web and docs, weigh the sources, then write a cited report.
no license 2 installs ★ 6 updated 2026-10-19
npx skills add lv416e/dotfiles --skill deep-research -a claude-codeWhat deep-research does
The skill gives the agent a procedure instead of a one-shot answer. Scoping comes first: questions to answer, boundaries, what counts as done. Then gathering, starting with a broad web sweep, narrowing to sub-topics, pulling library docs through Context7 and checking GitHub and forums for hands-on reports.
Analysis follows. Author, publication date, facts against opinions, whether separate sources land on the same numbers. Output has a fixed layout: Executive Summary, Detailed Findings, Critical Analysis, Practical Implications, Sources, Research Metadata. Extra guidance covers AI/ML work, breaking news, technical comparisons and business strategy.
It runs on the client's own tools: WebSearch, WebFetch, Context7 and the Explore subagent. No search tool in the agent means the gathering stage never starts. A report without the closing metadata block — queries used, source count, confidence level — is a sign the skill did not load.
What it helps with
- Break a broad topic into answerable sub-questions
- Compare two stacks using docs, GitHub issues and benchmarks
- Check whether several outlets agree on a trend
- Separate reporting from opinion in breaking news
- Get a cited report with confidence levels, not a link dump
Example prompts for the agent
- Do a deep research pass on AI agent evaluation metrics, with dated sources and a confidence level.
- Investigate whether we should switch frameworks — check the docs, GitHub discussions and benchmarks.
- Gather detailed information on recent MCP news and flag where the sources disagree.
How to install deep-research
1Skills CLI (Vercel)
npx skills add lv416e/dotfiles --skill deep-research -a claude-code2Manually: put the files in place
~/.claude/skills/deep-research/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/deep-research/
1Skills CLI (Vercel)
npx skills add lv416e/dotfiles --skill deep-research -a codex2Manually: put the files in place
.agents/skills/deep-research/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add lv416e/dotfiles --skill deep-research -a cursor2Manually: put the files in place
.cursor/skills/deep-research/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install lv416e/dotfiles dot_claude/skills/deep-research2Manually: put the files in place
.github/skills/deep-research/SKILL.md1Skills CLI (Vercel)
npx skills add lv416e/dotfiles --skill deep-research -a gemini-cli2Platform CLI
gemini skills install https://github.com/lv416e/dotfiles --path dot_claude/skills/deep-researchManually: put the files in place
.devin/skills/deep-research/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add lv416e/dotfiles --skill deep-research -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 deep-research
How do I install deep-research in Claude Code?
Run npx skills add lv416e/dotfiles --skill deep-research -a claude-code. The skill file then sits at ~/.claude/skills/deep-research/SKILL.md.
What does deep-research do?
It hands the agent a research routine: split the topic, gather material from more than one kind of source, judge the evidence, then write the report in a set layout. The topic comes from your request.
Which tools does the skill rely on?
Web search, WebFetch for specific pages, Context7 for library documentation and the Explore subagent for multi-step checks. Those tools have to be available in your client; the skill only sets the order of work.
Similar skills
All skills for Agent setup: memory, context, orchestration →deep-research by other authors comes first.