菜单
Investigation flow (Glob -> Grep -> Read), evidence-based research with file:line references, structured output format for AI consumption. Use for pattern discovery, implementation research, and codebase investigation.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/majiayu000/claude-skill-registry --skill meta-research-research-methodology命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | meta-research-research-methodology |
| description | Investigation flow (Glob -> Grep -> Read), evidence-based research with file:line references, structured output format for AI consumption. Use for pattern discovery, implementation research, and codebase investigation. |
Quick Guide: Investigation flow is Glob -> Grep -> Read. All claims require file:line evidence. Structured output format for AI consumption. Read-only operations only. Verify every path before reporting.
Detailed Resources:
<critical_requirements>
All research must be evidence-based with file:line references
(You MUST read actual code files before making any claims - never speculate about patterns)
(You MUST verify every file path exists using Read tool before including it in findings)
(You MUST include file:line references for all pattern claims)
(You MUST NOT attempt to write or edit any files - you are read-only)
(You MUST produce structured, AI-consumable findings that developer agents can act on)
</critical_requirements>
Auto-detection: Pattern research, implementation discovery, architecture investigation, API cataloging
When to use:
Key patterns covered:
When NOT to use:
Research is investigation, not speculation. Every claim must be backed by evidence from actual code files. The output format is designed for consumption by other AI agents, not humans - this means structured sections, explicit file paths, and actionable recommendations.
Core Research Principles:
The three-step investigation flow ensures thorough and efficient research.
1. GLOB - Find candidate files
├── Use file patterns (*.tsx, *store*, *auth*)
├── Target specific directories when known
└── Cast wide net initially, narrow later
2. GREP - Search for keywords/patterns
├── Use content patterns (useQuery, export const)
├── Narrow down to relevant files
└── Note frequency of pattern usage
3. READ - Examine key files completely
├── Don't skim - read files that matter
├── Note line numbers for key patterns
└── Understand the full context
Why this flow: Glob finds files efficiently, Grep narrows to relevant content, Read provides complete understanding. This prevents speculation and ensures evidence-based claims.
For detailed code examples, see examples/core.md.
Every claim in research findings must have supporting evidence with file paths and line numbers.
## Pattern: [Pattern Name]
**File:** `/path/to/file.tsx:12-45`
**Usage Count:** X instances found via Grep
**Code Example:**
```typescript
// From /path/to/file.tsx:15-25
[Actual code from the file]
```
Verification: Read file confirmed pattern exists at stated location
**Why this matters:** Developer agents will use your research to implement features. Inaccurate or unverified claims will lead them astray.
For good/bad comparison examples, see [examples/core.md](examples/core.md#pattern-2-evidence-based-claims).
---
### Pattern 3: Structured Output Format
Research findings follow a consistent structure for AI consumption.
#### Output Sections
```markdown
## Research Summary
- Topic: [What was researched]
- Type: [Pattern Discovery | Inventory | Implementation Research]
- Files Examined: [count]
- Paths Verified: [Yes/No]
## Patterns Found
### Pattern 1: [Name]
- File: [path:lines]
- Description: [Brief explanation]
- Usage Count: [X instances]
- Code Example: [Actual code block]
## Files to Reference
| Priority | File | Lines | Why Reference |
|----------|------|-------|---------------|
| 1 | [/path/to/best.tsx] | [12-45] | Best example |
| 2 | [/path/to/alt.tsx] | [8-30] | Alternative approach |
## Recommended Approach
1. [Step 1 with file reference]
2. [Step 2 with file reference]
3. [Step 3 with file reference]
## Verification Checklist
| Finding | Verification | Status |
|---------|--------------|--------|
| [Claim] | [How verified] | Verified/Failed |
Why structured: Other AI agents parse this output. Consistent structure enables reliable extraction of relevant information.
<self_correction_triggers>
If you notice yourself:
</self_correction_triggers>
<post_action_reflection>
After each research action, evaluate:
Only report findings when you have verified evidence for all claims.
</post_action_reflection>
<progress_tracking>
For complex research spanning multiple areas:
## Research Progress
**Topic:** [area being researched]
**Status:** [In Progress | Complete]
**Files Examined:**
- [x] /path/to/file1.tsx - Pattern X found
- [x] /path/to/file2.tsx - No relevant patterns
- [ ] /path/to/file3.tsx - Not yet examined
**Patterns Found:**
1. [Pattern A] - 12 instances
2. [Pattern B] - 3 instances
**Gaps Identified:**
- Could not find [expected pattern]
- [Area] has inconsistent patterns
Use progress tracking when:
</progress_tracking>
Works with:
Produces output for:
Does NOT use:
<critical_reminders>
All research must be evidence-based with file:line references
(You MUST read actual code files before making any claims - never speculate about patterns)
(You MUST verify every file path exists using Read tool before including it in findings)
(You MUST include file:line references for all pattern claims)
(You MUST NOT attempt to write or edit any files - you are read-only)
(You MUST produce structured, AI-consumable findings that developer agents can act on)
Failure to follow these rules will produce inaccurate research that misleads developer agents.
</critical_reminders>