AgentSkillsCN

tech-integration-research

快速在线调研方法,用于探索技术的整合方式、当前的最佳实践,以及配置要求。

SKILL.md
--- frontmatter
name: tech-integration-research
description:
  Rapid online research methodology for discovering how technologies integrate,
  current best practices, and configuration requirements.

Technology Integration Research

A focused online research methodology for answering specific technical questions about how tools, frameworks, and services work together.

When to Use

Use this skill when you need to:

  • Understand how Tool A integrates with Tool B
  • Find migration patterns from one technology to another
  • Discover configuration requirements and setup procedures
  • Learn about API usage and authentication flows
  • Identify common pitfalls and recommended approaches
  • Research current best practices (2024-2026)

Key indicator: Questions containing "How does X work with Y?" or "What do I need to set up Z?"

Research Methodology

1. Context Gathering

  • Check project documentation for current technology stack
  • Note versions of tools already in use
  • Understand what's changing (migration, new integration, upgrade)
  • Identify specific question to answer

2. Search Strategy

Official Documentation First:

  • Tool A official docs
  • Tool B official docs
  • Official integration guides (if they exist)

Current Practices:

  • Recent blog posts (2024-2026)
  • GitHub issues and discussions
  • Migration guides and changelogs
  • Example repositories or starter templates

Search Query Patterns:

code
"[Tool A] [Tool B] integration 2026"
"[Tool A] with [Tool B] setup guide"
"migrate from [Old Tool] to [New Tool] 2026"
"[Tool A] authentication [specific auth system]"
"[Tool A] Docker Compose example"
"[Tool A] version [X.x] [Tool B]"

3. Source Validation (Ongoing)

Prioritization:

  1. Official documentation (highest authority)
  2. Official blog posts from project maintainers
  3. Well-maintained community guides with recent updates
  4. GitHub discussions with maintainer responses
  5. Individual blog posts (verify date and version)

Confidence Levels: Assign confidence to each finding based on source quality and convergence:

  • High: Official docs confirm, or 3+ independent sources agree
  • Medium: 1-2 reliable sources, or official docs are ambiguous
  • Low: Single community source, or sources conflict

Red Flags:

  • Information older than 2 years (unless researching legacy)
  • Blog posts without version numbers
  • Contradictory information without explanation
  • Examples that don't show imports/dependencies
  • Tutorials that skip authentication/security

4. Information Synthesis (5 minutes)

Extract and organize:

  • Key integration patterns discovered
  • Required configuration (env vars, config files)
  • Dependencies to install
  • Step-by-step setup if available
  • Common pitfalls mentioned across sources
  • Gaps in documentation or open questions

Output Format

Structure findings for easy scanning and comparison:

markdown
## Question Researched

[Restate the specific question clearly]

## Summary

[2-3 sentence overview of key findings]

## Key Findings

- **[High/Medium/Low]** Finding 1: [Concise statement with source]
- **[High/Medium/Low]** Finding 2: [Concise statement with source]
- **[High/Medium/Low]** Finding 3: [Concise statement with source]

## Official Documentation

- [Tool Name Docs](URL) - What it covers
- [Integration Guide](URL) - What it covers

## Integration Pattern

[Code snippet or config example showing the integration]

```language
// Example code from official docs
```

Setup Requirements

Dependencies:

  • package-name@version - What it's for

Environment Variables:

  • VAR_NAME - Purpose and example value

Services Needed:

  • Service name - How it's used

Configuration Example

[Actual config file example if found]

Common Pitfalls

  • Issue 1: What goes wrong and how to avoid it
  • Issue 2: What goes wrong and how to avoid it

Version Compatibility

  • Tool A: version range tested
  • Tool B: version range tested
  • Known issues: [list any]

Open Questions / Gaps

  • What couldn't be found or remains unclear
  • What needs further investigation

Recommended Next Steps

  1. [Concrete action to take based on findings]
  2. [Next action]
  3. [etc.]

Sources Consulted

  1. Source Title - Published/Updated: [date], Accessed: [date]
  2. Source Title - Published/Updated: [date], Accessed: [date]
  3. [etc.]
code

## Quality Checklist

Before returning findings:
- [ ] Checked 3+ sources minimum
- [ ] Included at least one official source
- [ ] Assigned confidence levels (High/Medium/Low) to each finding
- [ ] Noted any conflicts between sources
- [ ] Included code/config examples (not just prose)
- [ ] Verified information is current (2024-2026)
- [ ] Listed all sources with URLs and publication dates
- [ ] Identified gaps or unclear areas
- [ ] Included concrete recommended next steps
- [ ] Structured output in standard format

## Example Questions

**Good (specific and actionable):**
- "How does Tool A authenticate with Tool B's JWKS endpoint?"
- "What's the recommended Docker Compose setup for Service X with Database Y?"
- "How do I migrate from Library A to Library B?"

**Too Broad (needs scoping):**
- "How does Tool X work?" → Scope to specific aspect
- "Tell me about Technology Y" → What specifically about it?
- "Best practices for category Z" → Too vague, narrow the question

## Scaling Effort to Question Complexity

Match your research effort to the question's complexity:

**Simple question** (e.g., "What package integrates X with Y?"):
- Check official docs for both tools
- Find 2-3 community examples or guides
- Stop when sources converge on the same answer

**Moderate question** (e.g., "How do I set up X with Y and Z?"):
- Check official docs for all tools involved
- Find 5-8 sources across different angles (setup guides, examples, discussions)
- Look for complete working examples
- Stop when you have a clear integration pattern and setup steps

**Complex question** (e.g., "How do I migrate from A to B while maintaining C?"):
- Check official migration guides and docs
- Find 10+ sources including GitHub discussions, migration stories, and examples
- Look for edge cases and common pitfalls
- Compare multiple approaches if they exist
- Stop when you have clear recommendations with trade-offs explained

**Too complex?** If you've checked 15+ sources and still need more depth, or the question involves architectural decisions beyond simple integration, consider using the investigator agent instead.