Standards and methods for conducting research, gathering information, and presenting findings with proper source attribution. Use when researching topics, analyzing documentation, or synthesizing information from multiple sources.
Resources
1Install
npx skillscat add molcajeteai/plugin/research-methods Install via the SkillsCat registry.
Research Methods
Standards for conducting research, evaluating sources, and presenting information in clear, well-formatted responses.
When to Use This Skill
Use this skill when:
- Researching documentation for tools, frameworks, or APIs
- Gathering information from multiple sources
- Synthesizing findings into coherent responses
- Creating reference materials
- Answering questions requiring external information
- Building knowledge bases
Key Principles
- Ask before researching - Clarify ambiguous queries before starting
- Prioritize official sources - Documentation over tutorials, primary over secondary
- Track all sources - Maintain attribution for every piece of information
- Match format to need - Simple questions get simple answers, complex get detailed
- Accuracy over speed - Cross-reference critical information
- User-focused presentation - Scannable, organized, actionable
Response Type Selection
Simple Response
Use when:
- Single, specific question
- How-to query with clear scope
- Quick reference lookup
- Command syntax or parameter info
- Definition or concept explanation
- User wants fast answer
Template:
# [Concise Title]
[Direct answer to the question]
[Code example if applicable]
## Sources
- [URL] - [Description]Example questions:
- "How do I amend a git commit?"
- "What is the Stop hook in Claude Code?"
- "How to use WebFetch tool?"
Detailed Response
Use when:
- Broad, exploratory question
- Request mentions "all", "detailed", "in-depth"
- Time estimate provided ("5 min intro", "quick overview")
- Multiple related concepts
- Comparative analysis
- Building reference material
Template:
# [Descriptive Title]
[Executive summary - 2-3 sentences providing overview]
## Overview
[Context and background]
## [Topic Section 1]
[Detailed information]
## [Topic Section 2]
[Detailed information]
## Key Takeaways
- [Point 1]
- [Point 2]
- [Point 3]
## Sources
- [URL] - [Description]Research Document Format
For deep research, produce a long-form document with the following structure:
---
date: YYYY-MM-DD
query: <original research input>
stack: <detected tech stack>
---
# Research: <Topic>
## Summary
{3-5 sentence answer — what this is, why it matters, what to use}
## Tech Stack Context
{Detected stack and how it affects the recommendations below}
## Key Findings
{Bulleted findings with source tier labels: [Tier 1], [Tier 2], [Tier 3], [Tier 4]}
## Library / Tool Comparison
{Table: name | what it does | stars/downloads | license | when to use}
## How-To: Implementation Guide
{Step by step, as if teaching someone from scratch}
### Step 1: ...
### Step 2: ...
{Code examples in the project's detected language at every step}
## Scenarios
{Table or subsections covering: basic case, edge cases, production considerations}
## Diagrams
{Mermaid flowcharts, sequence diagrams, or architecture diagrams as applicable}
## Knowledge Gaps
{What this research didn't cover, where to look next}
## Sources
{Tiered list: Tier 1 (Authoritative) > Tier 2 (Secondary) > Tier 3 (Community) > Tier 4 (Unverified)}Tech Stack Detection
Scan the project root for these indicators to determine the tech stack. Multiple matches are additive.
| File | Stack |
|---|---|
README.md |
Project documentation (read for context) |
go.mod |
Go |
Cargo.toml |
Rust |
pyproject.toml / setup.py |
Python |
package.json + tsconfig.json |
TypeScript/Node |
package.json (no tsconfig) |
JavaScript/Node |
next.config.* |
Next.js (React) |
vite.config.* |
Vite (React/Vue/Svelte) |
Makefile only |
C/C++ or mixed |
| none found | Language-agnostic |
Use the detected stack to:
- Provide code examples in the correct language
- Search the appropriate package registry (npm, pkg.go.dev, PyPI, crates.io)
- Reference stack-specific conventions and patterns
Research Process
1. Clarification Phase
Before starting research:
- Identify ambiguities in the query
- Ask specific clarifying questions
- Don't assume user intent
- Use AskUserQuestion tool for multiple clarifications
2. Source Selection
Priority order:
Tier 1: Authoritative Primary Sources
- Official documentation from tool/framework creators
- Published specifications and standards
- Official repositories and codebases
- First-party blog posts and announcements
Tier 2: Authoritative Secondary Sources
- Well-known educators and tutorial sites
- Official community resources
- Verified contributor blogs
- Respected technical publications
Tier 3: Community Sources
- Stack Overflow answers
- GitHub issues and discussions
- Developer blogs
- Tutorial websites
Tier 4: Unverified Sources
- Content farms, unattributed tutorials, outdated resources
- Use with extreme caution, always verify against higher-tier sources
3. Information Gathering
For each source:
- Read thoroughly
- Extract relevant information
- Note source URL and description
- Identify key points
- Check publication date/relevance
Cross-referencing:
- Verify critical facts across 2+ sources
- Note version-specific information
- Flag contradictions for investigation
- Prefer recent over outdated
4. Synthesis
Organize information:
- Group related concepts
- Order from general to specific
- Identify patterns and relationships
- Extract key takeaways
Quality checks:
- Is it accurate?
- Is it complete for the query?
- Is it clearly presented?
- Are sources properly attributed?
Formatting Standards
Markdown Structure
Headers:
- H1 (
#) - Title only - H2 (
##) - Major sections - H3 (
###) - Subsections if needed - Keep hierarchy shallow (max 3 levels)
Code blocks:
# Always specify language
# Include comments for clarityLists:
- Use bullets for unordered items
- Use numbers for sequential steps
- Keep items parallel in structure
- One idea per bullet point
Emphasis:
- Bold for key terms and important points
- Italic sparingly for subtle emphasis
Code fontfor technical terms, commands, file names
Source Attribution
Format:
## Sources
- [URL] - [Brief description of what info came from this source]Best practices:
- List in order of importance/relevance
- Keep descriptions concise (5-10 words)
- Include official docs first
- Don't duplicate similar sources
Quality Standards
Accuracy
- Cross-reference critical information
- Note version-specific details
- Flag assumptions or uncertainties
- Prefer official sources
Completeness
- Answer the actual question asked
- Include necessary context
- Provide examples when helpful
- Cover edge cases if relevant
Clarity
- Use simple, direct language
- Define technical terms
- Organize information logically
- Make content scannable
Attribution
- Cite every source used
- Track where information came from
- Describe what each source contributed
- Link to original documentation
Error Handling
Insufficient Information
I found limited information about [topic]. Based on available sources:
[Present what was found]
This might indicate:
- Recent/unreleased feature
- Deprecated functionality
- Different terminology
Would you like me to search with alternative terms?Contradictory Sources
I found conflicting information:
- Source A: [X]
- Source B: [Y]
This appears to be due to [version/context/timing].
The most current information suggests: [recommendation]No Results
I couldn't find reliable sources for [topic].
Could you:
- Verify the terminology?
- Provide more context?
- Specify the tool/version?Related Files
references/templates.md- Detailed template examples with full samplesreferences/search-strategies.md- Advanced search techniques per domainreferences/source-evaluation.md- Criteria for assessing source quality