Leaderboard/Research/Deep-Research-skills
Last commit on April 10, 2026·Created on December 29, 2025

Weizhena/Deep-Research-skills

Turns a research topic into a structured, multi-agent investigation.
Combined rank
#177
across all skills
In Research
#11
category rank
Stars
2.2k
+2.2% in last 7d
Forks
177
+2.3% in last 7d
Watchers
14
0.0% in last 7d
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
TL;DR

This skill collection guides structured research from outline to report. It generates an item and field framework, launches independent agents to research each item in parallel, then compiles a markdown report that covers all fields while marking uncertain values. The pipeline serves market analysts, product managers, and academic researchers who need systematic benchmarks, competitive analysis, or literature surveys with built-in human oversight at each stage.

WHO IT'S FOR
Market Research Analysts
conduct structured competitive analysis
Academic Researchers
perform preliminary literature and topic scoping
Technology Strategists
evaluate and compare technology options
Product Managers
research market trends and competitor products
Repository contents

20 skill files

Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.

Use this skill collection

Follow the documented setup, then try a first task.

I’m using

Claude Code — English bundle

Installs: Installs the complete English research-en skill bundle plus the required web-search agent and modules. · Claude Code

Before you start
  • Git
  • Python with pip
  • Claude Code 2.1.0+ for direct /skill-name triggers; older versions use run /skill-name
  • The destination directories ~/.claude/skills/ and ~/.claude/agents/ must exist.

Run from: Clone from the desired parent directory; run all subsequent steps from the deep-research-skills repository root in the same terminal.

  1. In the directory where the repository should be created, clone it and enter the repository.

    git clone https://github.com/Weizhena/deep-research-skills.git
    cd deep-research-skills
  2. Install the complete English skill bundle into Claude Code.

    cp -r skills/research-en/* ~/.claude/skills/
  3. Install the required web-search agent and its modules.

    cp agents/web-search-agent.md ~/.claude/agents/
    cp -r agents/web-search-modules ~/.claude/agents/
  4. Install the required Python dependency.

    pip install pyyaml
README.md · Checked Sep 18, 2026

Give it something to do.

Author’s example

Create an initial research outline

Uses research

  • Research topic; replace the sample topic “AI Agent Demo 2025” with the topic you want to investigate
  • Desired time range when the skill asks for it
  • Any existing field-definition file, if available
  • Preferred batch size and items per agent when prompted

Check the repository for this skill’s setup.

/research AI Agent Demo 2025
skills/research-zh/research/SKILL.md · Checked Sep 18, 2026README.md · Checked Sep 18, 2026
Suggested first task

Add named objects to an existing outline

Uses research-add-items

  • [PROJECT]: project directory beneath the current working directory
  • [ITEM NAMES]: one or more specific research objects to add
  • An existing [PROJECT]/outline.yaml file

Check the repository for this skill’s setup.

Use the research-add-items skill in the current working directory containing [PROJECT]/outline.yaml. Add [ITEM NAMES] as research objects, avoid duplicates, and show the proposed update for confirmation before saving it.
Suggested first task

Add field definitions to an existing outline

Uses research-add-fields

  • [PROJECT]: project directory beneath the current working directory
  • [FIELD NAMES AND DESCRIPTIONS]: the fields to consider adding
  • An existing [PROJECT]/fields.yaml file

Check the repository for this skill’s setup.

Use the research-add-fields skill in the current working directory containing [PROJECT]/fields.yaml. Propose fields from [FIELD NAMES AND DESCRIPTIONS], let me choose their categories and detail_level values, show the proposed additions for confirmation, and then save the confirmed fields.