Files

T

jilinchenandClaude Opus 4.5 302f2733d3 refactor: separate outline.yaml and fields.yaml

- Output to {topic}/ directory structure
- fields.yaml now independent file for field definitions
- outline.yaml contains items + execution config
- Deep research agents read fields.yaml independently (hard constraint)
- Update README with new directory structure

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

2025-12-30 10:42:58 +08:00

2.4 KiB

Raw Blame History

Deep Research Skill for Claude Code

Inspired by RhinoInsight: Improving Deep Research through Control Mechanisms for Model Behavior and Context

A structured research workflow skill for Claude Code, supporting two-phase research: outline generation (extensible) and deep investigation. Human-in-the-loop design ensures precise control at every stage.

Use Cases

Academic Research: Paper surveys, benchmark reviews, literature analysis
Technical Research: Technology comparison, framework evaluation, tool selection
Market Research: Competitor analysis, industry trends, product comparison
Due Diligence: Company research, investment analysis, risk assessment

Commands

Command	Description
`/research`	Generate research outline with items and fields
`/research/add-items`	Add more items to existing outline
`/research/add-fields`	Add more fields to existing outline
`/research/deep`	Deep research each item with parallel agents

Installation

Copy the skills/ folder to your Claude Code directory:

cp -r skills/* ~/.claude/skills/

Workflow

Phase 1: Generate Outline

/research <topic>

Uses model knowledge + web search
Asks for existing field definitions
Creates {topic}/ directory with separated files

Phase 2: Deep Research

/research/deep

Reads outline and fields automatically from current directory
Launches parallel agents (5 per batch)
Agents read fields.yaml independently (not passed in prompt)
Outputs structured JSON per item
Supports resume from checkpoint

Optional: Expand Outline

/research/add-items    # Add more research targets
/research/add-fields   # Add more field definitions

Output Format

Directory Structure

{topic}/
  ├── outline.yaml    # items + execution config
  ├── fields.yaml     # field definitions
  └── results/        # deep research outputs

outline.yaml

topic: "your topic"
items:
  - name: "Item1"
    source: "source info"
execution:
  batch_size: 5

fields.yaml

basic_info:
  - name: "field_name"
    description: "field description"
    detail_level: "detailed|brief"

Research Result (JSON)

Each item outputs a structured JSON file with all defined fields.

References

RhinoInsight: Improving Deep Research through Control Mechanisms for Model Behavior and Context

License

MIT

2.4 KiB Raw Blame History