AgentSkillsCN

tech-trend-researcher

当用户请求研究技术主题、技术趋势、行业发展或任何需要网络搜索、来源分析与结构化文档的主题时，应使用此技能。它提供系统化的工作流，从多个网络来源收集信息，合成结论为全面摘要，并保存结果到本地文件。

SKILL.md
--- frontmatter
name: tech-trend-researcher
description: This skill should be used when users request research on technical topics, technology trends, industry developments, or any subject requiring web search, source analysis, and structured documentation. It provides a systematic workflow for gathering information from multiple web sources, synthesizing findings into a comprehensive summary, and saving results to a local file.
license: Complete terms in LICENSE.txt

Tech Trend Researcher

This skill provides a systematic workflow for researching technical topics, analyzing trends, and documenting findings in a structured format.

Purpose

Research technical topics or trends by:

  1. •Conducting targeted web searches
  2. •Analyzing multiple authoritative sources
  3. •Synthesizing information into comprehensive summaries
  4. •Saving structured documentation to local files

This skill solves the problem of ad-hoc research that lacks structure, often resulting in incomplete coverage or poor documentation quality.

When to Use This Skill

Use this skill when users request:

  • •Technology trend analysis (e.g., "Research TypeScript trends")
  • •Industry developments (e.g., "Latest AI advancements")
  • •Framework/library updates (e.g., "What's new in React 19")
  • •Technical topic overviews (e.g., "Kubernetes best practices 2025")
  • •Comparative analyses (e.g., "Python vs Go for microservices")
  • •Market research on technical products or tools

Trigger phrases include:

  • •"Research [topic] and save to file"
  • •"What are the latest trends in [technology]"
  • •"Summarize recent developments in [field]"
  • •"Create a report on [topic]"

Workflow

Follow this systematic four-phase workflow:

Phase 1: Initial Web Search

Conduct broad web search to identify key themes and authoritative sources.

Steps:

  1. •Formulate search query targeting latest information (include current/recent year when relevant)
  2. •Use web_search tool with appropriate max_results (typically 8-12)
  3. •Scan results to identify:
    • •Major themes and trends
    • •Authoritative sources (official docs, major tech publications, industry leaders)
    • •Specific statistics or data points
    • •Key terminology and concepts

Search Query Guidelines:

  • •Include temporal indicators: "2024", "2025", "latest", "recent"
  • •Add context keywords: "trends", "developments", "updates", "new features"
  • •Be specific but not overly narrow
  • •Examples:
    • •Good: "TypeScript latest trends 2024 2025"
    • •Good: "React 19 new features release"
    • •Too narrow: "TypeScript 5.3.2 specific bug fixes"
    • •Too broad: "programming languages"

Phase 2: Deep Source Analysis

Fetch and analyze 2-4 most relevant sources for detailed information.

Steps:

  1. •

    Select sources based on:

    • •Recency (prefer 2024-2025 content)
    • •Authority (official docs, major publications, recognized experts)
    • •Comprehensiveness (detailed articles over brief mentions)
    • •Relevance to user's specific question
  2. •

    Use fetch tool to retrieve full content:

    • •Set appropriate max_length (6000-10000 for detailed articles)
    • •Fetch multiple sources in parallel when possible
    • •Prioritize sources that provide different perspectives or complementary information
  3. •

    Extract key information:

    • •Specific features, versions, or capabilities
    • •Statistics, adoption rates, or metrics
    • •Expert opinions or industry consensus
    • •Code examples or technical details
    • •Future roadmap or predictions

Phase 3: Information Synthesis

Organize and synthesize information into a structured summary.

Structure Guidelines:

Always include these core sections:

  1. •Header: Title, date, sources
  2. •Overview: 2-3 paragraph executive summary
  3. •Main Trends/Topics: Numbered sections with clear headings
  4. •Key Features/Details: Bullet points or subsections
  5. •Market Context: Adoption rates, industry impact, statistics
  6. •Future Outlook: Predictions, roadmap, or upcoming changes
  7. •Conclusion: Key takeaways and recommendations
  8. •References: Source links or citations

Formatting Standards:

  • •Use clear hierarchy: Headers (===), subheaders (---), bullet points
  • •Include specific data: percentages, dates, version numbers
  • •Provide examples: code snippets, use cases, scenarios
  • •Add visual separation: line breaks, section dividers
  • •Use symbols for emphasis: checkmarks, bullets, arrows
  • •Keep paragraphs concise: 3-5 sentences maximum

Content Quality Standards:

  • •Accuracy: Verify claims across multiple sources
  • •Completeness: Cover all major themes identified in Phase 1
  • •Balance: Include both benefits and limitations/challenges
  • •Specificity: Avoid vague statements; provide concrete details
  • •Actionability: Include practical implications or recommendations

Phase 4: File Documentation

Save the synthesized summary to the user's specified file path.

Steps:

  1. •Confirm file path from user request (use absolute paths)
  2. •Use write_file tool with mode='rewrite'
  3. •Include the complete structured summary from Phase 3
  4. •For files over 50 lines, consider chunking (write first chunk, then append)

File Naming Conventions:

  • •Use descriptive names: typescript_trends.txt, react_19_features.md
  • •Include date if relevant: ai_trends_2025.txt
  • •Use underscores for spaces: kubernetes_best_practices.txt
  • •Prefer .txt or .md extensions for readability

Best Practices

Search Strategy

  • •Cast wide net first: Initial search should be broad to capture ecosystem
  • •Then go deep: Follow up with targeted searches if gaps remain
  • •Multiple queries: Consider 2-3 different search angles for comprehensive coverage
  • •Validate recency: Prioritize sources from past 6-12 months for trend analysis

Source Selection

  • •Official over community: Official docs and announcements trump blog posts
  • •Data over opinion: Sources with statistics or benchmarks are more valuable
  • •Multiple perspectives: Do not rely on single source; cross-reference
  • •Original over secondary: Link to primary sources when available

Synthesis Quality

  • •Structure matters: Well-organized content is more valuable than comprehensive but chaotic information
  • •Context is key: Explain why trends matter, not just what they are
  • •Be critical: Note limitations, controversies, or competing viewpoints
  • •Stay objective: Present balanced view even if sources are biased

Common Pitfalls to Avoid

  • •Over-reliance on first search: Do not stop after initial search if coverage is incomplete
  • •Ignoring dates: Outdated information can mislead users about current trends
  • •Too much detail: Focus on signal over noise; summarize rather than transcribe
  • •Poor structure: Wall of text reduces value; use hierarchy and formatting
  • •Missing context: Statistics without context are meaningless
  • •No conclusion: Always provide synthesis and takeaways

Output Quality Checklist

Before saving the file, verify:

  • •All major themes from initial search are covered
  • •At least 2-3 authoritative sources are cited
  • •Specific data points (dates, versions, percentages) are included
  • •Structure follows standard format with clear sections
  • •Content is well-formatted with headers and bullets
  • •Conclusion provides actionable insights
  • •File path is correct and absolute
  • •No placeholder text or TODOs remain

Examples

Example 1: Technology Trend Research

User Request: "Research TypeScript trends and save to /Users/user/Desktop/typescript_trends.txt"

Workflow Execution:

  1. •Search: "TypeScript latest trends 2024 2025" with 10 results
  2. •Fetch: 2 detailed articles from authoritative sources
  3. •Synthesize: 10-section structured summary covering adoption rates, new features, ecosystem changes
  4. •Save: Complete formatted document to specified path

Key Success Factors:

  • •Included specific statistics (12% to 35% adoption growth)
  • •Covered multiple dimensions (features, tooling, ecosystem)
  • •Provided actionable conclusion
  • •Well-structured with clear hierarchy

Example 2: Framework Update Research

User Request: "What's new in React 19? Save summary to ~/Documents/react19.md"

Workflow Execution:

  1. •Search: "React 19 new features release 2024"
  2. •Fetch: Official React blog post plus community analysis
  3. •Synthesize: Feature-by-feature breakdown with code examples
  4. •Save: Markdown format with syntax highlighting

Example 3: Comparative Analysis

User Request: "Compare Rust vs Go for web services and document findings"

Workflow Execution:

  1. •Search: "Rust vs Go web services 2024 comparison"
  2. •Fetch: Multiple benchmark articles and developer surveys
  3. •Synthesize: Side-by-side comparison with use case recommendations
  4. •Save: Structured comparison table with detailed sections

Adaptation for Different Domains

This workflow can be adapted for non-technical research:

Business/Market Research:

  • •Search industry reports and market analyses
  • •Focus on market size, growth rates, key players
  • •Include competitive landscape and trends

Policy/Regulatory Research:

  • •Search official government or regulatory sources
  • •Focus on timeline, requirements, implications
  • •Include compliance considerations

Scientific/Academic Research:

  • •Search academic publications and research papers
  • •Focus on methodologies, findings, citations
  • •Include limitations and future research directions

Notes

  • •Objectivity: Present balanced view; avoid promotional language
  • •Timeliness: Always check publication dates; trends evolve quickly
  • •User's Language: If user requests in non-English, consider translating summary
  • •File Permissions: Verify user has write access to specified path
  • •Follow-up: If user requests clarification, be prepared to expand specific sections