AugmentClaude
User ResearchETL

Deep Research

By feiskyer· feiskyer/claude-code-settings· 0

Orchestrate parallel research tasks and synthesize findings into a polished report.

Installation

  1. Make sure Claude is on your device and in your terminal.

    Skills load from ~/.claude/skills/ when Claude Code starts up — so you need it on your machine first. If you don't have it yet, install it once with the command below, then run claude in any terminal to verify.

    One-time setup
    npm i -g @anthropic-ai/claude-code

    Already have it? Skip ahead.

  2. Paste into Claude Code or into your terminal.
    Install
    Sign up to copy
    Sign up free to reveal

    This copies the whole skill folder into ~/.claude/skills/deep-research-feiskyer/ — the SKILL.md plus any scripts, reference docs, or templates the skill ships with. Safe default: works for every skill.

    Faster alternative (instruction-only skills)

    Skips the clone and grabs only the SKILL.md file. Don't use this if the skill ships Python scripts, reference markdowns, or asset templates — they won't be downloaded and the skill will fail when it tries to load them.

    Quick install (SKILL.md only)
    Sign up to copy
    Sign up free to reveal
  3. Restart Claude Code.

    Quit and reopen Claude Code (or any other agent that loads from ~/.claude/skills/). New skills are picked up on startup.

  4. Just ask Claude.

    Skills auto-activate when your request matches the skill's description — no slash command needed. Trigger phrases live in the skill's own frontmatter; you can read them in the “What this skill does” section above.

Prefer to read the source first? Open on GitHub.

When Claude uses it

Use for systematic research jobs — splits a research goal into parallel sub-goals, runs each in a headless Claude subprocess, and aggregates everything into a polished report. Good for web research, competitive analysis, and long-form evidence synthesis.

What this skill does

This skill orchestrates large-scale research projects by breaking them into parallel sub-tasks, running each in independent Claude processes, and synthesizing results into a polished final report. It handles systematic research goals across web research, competitive analysis, and evidence synthesis by splitting work across multiple subprocess runs, then aggregating and refining outputs into a structured, standalone document.

What it does

  • Decomposes research goals into parallel sub-tasks (by topic, time period, data slice, or other logical dimension), with real sampling and discovery before execution begins.
  • Launches parallel subprocesses using claude -p for each sub-task, with fine-grained tool permissions via --allowedTools to keep tasks contained and efficient.
  • Orchestrates tool priority: skills first, then MCP tools (preferring firecrawl, then exa), then WebFetch/WebSearch as fallback—all configured per subprocess.
  • Aggregates outputs programmatically from child processes into a structured base draft, then refines it chapter-by-chapter into a polished final report.
  • Maintains full auditability: logs execution progress in real time, caches raw data, and stores all artifacts under .research/<name>/ with traceable references and source links.
  • Enforces file-based delivery: final report lands as a standalone document (not pasted into chat), with internal working files and intermediate outputs organized by directory.

How to use it

Provide a research objective. The skill first clarifies your goal, identifies dimensions to parallelize, and runs minimal live sampling to confirm data availability—showing you real results before committing to full execution. After you confirm ("execute," "go," or similar), it spins up subprocesses, monitors them in parallel, aggregates their findings, then refines the combined output into a polished report. You hand it the research question; it returns the file path to your completed report along with a summary of key findings.

Note: this skill's internal instructions are written in Chinese. Claude reads them natively and will work with you in English.

Related skills

C

Co-Marketing Partnerships

coreyhaines31

Find ideal partners and plan joint marketing campaigns with other companies.

I

Idea Sprint

solanabr

Validate crypto startup ideas and decide whether they're worth building.

J

Jobs to be Done Canvas

product-on-purpose

Map customer motivations across functional, emotional, and social dimensions.

Apache-2.0
A

App Store Review Arbitrage

Varnan-Tech

Turn competitor reviews into marketing positioning and ad copy opportunities.