name: deep-research
description: Web research with Graph-of-Thoughts for fast-changing topics. Use when user requests research, analysis, investigation, or comparison requiring current information. Features hypothesis testing, source triangulation, claim verification, Red Team, self-critique, and gap analysis. Supports Quick/Standard/Deep/Exhaustive tiers. Creative Mode for cross-industry innovation.

Deep Research

Enhanced research engine for topics where training data is outdated.

Quick Start

Standard Mode

CLASSIFY → LANDSCAPE SCAN → SCOPE → HYPOTHESIZE → PLAN → [PLAN PREVIEW*] → RETRIEVE
→ GAP ANALYSIS → TRIANGULATE → SYNTHESIZE → RED TEAM → SELF-CRITIQUE → PACKAGE

*Deep+ tier only

LANDSCAPE SCAN (MANDATORY - Before Anything Else)

[Search for OVERVIEW first - NO known entity names in query!]
WebSearch: "[topic] landscape overview [current year]"
WebSearch: "top [topic] list [current year]"
WebSearch: "[topic] ecosystem players [current year]"

❌ WRONG: "DeepSeek Qwen performance 2025" (uses names you already know)
✅ RIGHT: "China open source LLM models list 2025" (discovers what exists)

→ Extract ALL entity names from results
→ List: Discovered (new to you) vs Confirmed (you knew)
→ THEN proceed to SCOPE with complete picture

Why: You cannot research what you don't know exists. Scan the landscape FIRST.

Creative Mode

ABSTRACT → MAP (3-5 domains) → SEARCH → GENERALIZE → SYNTHESIZE

Trigger: "creative mode", "cross-industry", "what do others do"

Classification

Intensity Tiers

Core Rules

Parallel Search (MANDATORY)

[Single message]
WebSearch: "[topic] 2025"
WebSearch: "[topic] limitations"
WebSearch: "[topic] vs alternatives"

Claim Types

Confidence Format (C1 claims)

**Claim:** [Statement]
**Confidence:** HIGH/MEDIUM/LOW
**Reason:** [Why]
**Sources:** [1][2]

Anti-Hallucination

• Every C1 cites [N] immediately
• Use "According to [1]..."
• Admit: "No sources found for X"

URL Fallback

If WebFetch returns 403:

curl -s --max-time 60 "https://r.jina.ai/https://example.com"

GitHub Repository Research

When research reveals interesting GitHub repositories that could provide deeper insights:

1. ASK user before cloning:
   "เจอ repo ที่น่าสนใจ: [repo-name] — ต้องการให้ clone มาศึกษา code โดยละเอียดไหมคะ?"

2. If user agrees, clone to study folder:
   bash cd /mnt/d && git clone [repo-url] [repo-name]-study

3. Key files to read:

4. package.json / pyproject.toml — dependencies, entry points
5. src/index.ts or src/main.py — main logic
6. src/types/ — data structures
7. README.md — overview

Why: Cloned repos allow deeper code analysis than WebFetch summaries.

Finding Details in References

Scripts

Output File (MANDATORY)

After completing research, ALWAYS save to markdown file:

research/[topic-slug]-[YYYY-MM-DD].md

Example: research/china-opensource-ai-2025-01-04.md

• Create research/ folder if it doesn't exist
• Why: Research takes effort. Save it for future reference.

Related Skills