deep-research · akillness — Claude Code skill: what it does, how to install
deep-research turns a topic into an extendable outline, sends parallel agents through each item, then writes a markdown report. Built for comparing a set of things across shared fields.
no license 120 installs ★ 61 updated 2026-10-24
npx skills add akillness/jeo-skills --skill deep-research -a claude-codeWhat deep-research does
The skill acts as a router. It sorts a request into one of four phases: outline, deep, report, web-search. Then it loads the matching instruction file from references/ and follows the steps: build an extendable outline for the topic, run a separate web search per item in parallel, write the findings as JSON with a field-coverage check, render a markdown report with a table of contents. Each phase stops at a checkpoint and waits for your go-ahead.
Four pipelines carry their own prompt templates and output contract. Outline can add items and fields, deep runs the per-item agents, report writes the TOC and the field-by-field text, web-search splits queries across five source modules: GitHub debugging, general web, academic papers, Chinese tech sources, Stack Overflow. Everything is plain Markdown, so it runs in Claude, Codex, Gemini CLI, Cursor, Copilot and OpenCode.
Deep and report need Python 3 with pyyaml, since validate_json.py will not check field coverage without it. On OpenCode, web search needs OPENCODE_ENABLE_EXA=1 set first. Ready-made slash commands for Claude, OpenCode and Codex live upstream and get copied by WITH_UPSTREAM=1 bash scripts/install.sh.
What it helps with
- draft a comparison outline for a set of tools or papers
- fan out per-item web searches in parallel
- validate that no outline field is left empty
- render a shareable markdown report with a TOC
- extend an existing outline with more items or fields
Example prompts for the agent
- Build a research outline for open-source vector databases with fields for licence, language and GPU support.
- Now go deep on every item and check that each field is filled.
- Generate
report.mdwith a table of contents from the collected JSON.
How to install deep-research
1Skills CLI (Vercel)
npx skills add akillness/jeo-skills --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 akillness/jeo-skills --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 akillness/jeo-skills --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 akillness/jeo-skills .agent-skills/deep-research/SKILL.md2Manually: put the files in place
.github/skills/deep-research/SKILL.md1Skills CLI (Vercel)
npx skills add akillness/jeo-skills --skill deep-research -a gemini-cli2Platform CLI
gemini skills install https://github.com/akillness/jeo-skills --path .agent-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 akillness/jeo-skills --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 akillness/jeo-skills --skill deep-research -a claude-code. The skill lands at ~/.claude/skills/deep-research/SKILL.md.
What does deep-research do?
It picks the right phase for your request (outline, deep, report, web-search) and follows that pipeline through to a markdown report.
What does it need to run?
Python 3 with pyyaml for the deep and report phases. On OpenCode, set OPENCODE_ENABLE_EXA=1 so web search works.
Similar skills
All skills for Agent setup: memory, context, orchestration →deep-research by other authors comes first.