Research to Study Guide
Transform any topic into a comprehensive, curated study guide through systematic research, quality validation, and audience-focused synthesis.
Inputs / Outputs
| What | Path | |
|---|---|---|
| Input | Topic and learner profile from user | Conversational |
| Working file | Research scratchpad (created by skill) | pipeline/scratchpad/[topic]-research-scratchpad.md |
| Output | Finished study guide | knowledge/[topic]-study-guide-v1.md |
| Prompts | Full prompt templates for this workflow | prompts/ subdirectory in this skill |
Quick Start
- Get the topic and learner profile from the user
- Create scratchpad at
/pipeline/scratchpad/[topic]-research-scratchpad.md - Execute research using web search
- Evaluate quality — iterate if gaps exist
- Synthesize into study guide at
/knowledge/[topic]-study-guide-v1.md
Phase 0: Setup
Ask the user (or infer from context):
topic: "[The subject to research]"
learner_profile:
role: "[Who is learning — job title, experience level]"
goal: "[What they want to achieve]"
context: "[Why now, what they'll use it for]"
target_sources: 50 # Adjust based on topic breadth
Audience Analysis (Do This First)
Before ANY research, answer:
| Question | Answer |
|---|---|
| WHO is learning this? | |
| WHAT do they care about vs noise? | |
| WHY do they need this? | |
| WHAT sources would they trust? | |
| WHAT should be excluded? |
Phase 1: Generate Research Questions
Create 3-5 questions per category:
Required Categories
-
Foundations (Critical)
- What is [topic] and why does it matter?
- Core concepts and vocabulary
- Mental models experts use
- Common misconceptions
-
Frameworks & Models (Critical)
- Established frameworks
- Processes practitioners follow
- Templates and tools
-
Key Practitioners (High)
- Recognized experts
- Their key contributions
- Where experts agree/disagree
-
Source Discovery (High)
- Essential books
- Blogs/newsletters
- Podcasts and videos
- Courses
-
Real Examples (High)
- Best documented examples
- Instructive failures
- Before/after transformations
-
Skills & Practice (Medium)
- Skills that distinguish great practitioners
- What to practice
- Learning sequence
Question Format
Each question needs:
- Clear query with context
- Success criteria (what makes it complete)
- Priority (critical/high/medium/low)
- Source requirements
Phase 2: Execute Research
Use web search to answer each question. For each:
## Q[N]: [Question]
### Findings
[What you discovered]
### Sources
| Source | Type | Quality (1-5) | Notes |
|--------|------|---------------|-------|
### Gaps
[What's missing]
Source Quality Rubric
| Score | Criteria |
|---|---|
| 5 | Authoritative expert, unique insights, practical |
| 4 | Respected source, solid content, actionable |
| 3 | Decent coverage, some value |
| 2 | Surface-level, generic, or dated |
| 1 | Low quality, unsupported claims |
Phase 3: Quality Checkpoint
Before synthesis, evaluate:
| Metric | Target |
|---|---|
| Foundations covered | Yes |
| Frameworks found | 3+ |
| Experts identified | 5+ |
| Sources discovered | 30+ |
| Examples documented | 5+ |
Quality Bands:
- Bad (0-0.4): Must iterate — critical gaps
- Acceptable (0.4-0.7): Can proceed with limitations
- Great (0.7-1.0): Proceed to synthesis
If gaps exist, generate follow-up questions and iterate (max 3 times).
Phase 4: Synthesis
Create study guide with audience filtering:
# [Topic] Study Guide
> **For:** [Learner profile]
> **Goal:** [What they'll achieve]
> **Time Investment:** [Estimated hours]
> **Last Updated:** [Date]
## How to Use This Guide
[Brief instructions]
## Part 1: Foundations
[Core concepts, mental models, misconceptions]
## Part 2: Frameworks & Process
[Key frameworks with attribution and examples]
## Part 3-4: [Topic-Specific Sections]
[Adapt based on topic]
## Part 5: Skills & Practice
[How to get good]
## Appendix A: Curated Source Library
[Best sources organized by type]
## Appendix B: Templates & Tools
[Ready-to-use resources]
## Appendix C: Learning Path
[Recommended sequence]
Anti-Hallucination Rules
- NEVER guess URLs — Only cite verified sources
- NEVER invent sources — No fabricated books, authors, or frameworks
- NEVER fake statistics — Say "not found" if you don't have data
- Trust web research — Don't flag recent info as dubious
- Cite everything — Every claim needs a source
File Outputs
| File | Purpose |
|---|---|
/pipeline/scratchpad/[topic]-research-scratchpad.md | Living research notes |
/knowledge/[subdirectory]/[topic]-study-guide-v1.md | Final study guide |
Subdirectory Routing
Place the final study guide in the subdirectory that matches its primary use case:
| Subdirectory | When to use | Examples |
|---|---|---|
content/ | Writing, style, content creation | writing craft, hooks, engagement |
building/ | Company ops, leadership, scaling | handbooks, operator frameworks |
product/ | Product strategy, engineering, technical topics | product leadership, architecture |
domain/ | Industry-specific or specialized knowledge | market research, competitive analysis |
If the topic doesn't clearly fit one subdirectory, ask the user which one to use.
Example Usage
User: "I want to become an expert in company handbooks"
- Create scratchpad:
/pipeline/scratchpad/company-handbook-research-scratchpad.md - Audience: Founders at 10-100 person companies
- Research: Public handbooks (Basecamp, GitLab, Valve, Netflix), structure patterns, expert advice
- Evaluate: 24 sources, 6 frameworks, 10+ examples = Great (0.85)
- Synthesize:
/knowledge/building/company-handbook-study-guide-v1.md
User: "Research [industry topic] for our team"
- Create scratchpad:
/pipeline/scratchpad/[topic]-research-scratchpad.md - Define audience from user context
- Research: Web search across trusted sources, follow citation trails
- Evaluate: Score against quality bands
- Synthesize:
/knowledge/domain/[topic]-study-guide-v1.md
Deep Reference
Full prompt template: research-to-study-guide-prompt-v1.md
