Agent Skills
› afx-team/hebb-mind
› research-survey
research-survey
GitHub针对指定主题执行结构化研究调查,结合学术论文、开源项目及行业报告进行综合检索与对比分析。最终生成包含摘要、学术/开源/行业趋势、差距分析及建议的Markdown格式研究报告。
触发场景
用户请求对某个技术主题进行学术研究综述
用户需要对比某领域的顶级论文与GitHub开源项目
用户希望获取包含行业洞察和差距分析的综合调研报告
安装
npx skills add afx-team/hebb-mind --skill research-survey -g -y
SKILL.md
Frontmatter
{
"name": "research-survey",
"arguments": [
{
"name": "topic",
"required": true,
"description": "The research topic to survey"
},
{
"name": "output",
"required": false,
"description": "Output file path for the survey report (default writes to repo_pages\/surveys\/)"
}
],
"description": "Conduct a structured research survey on a topic, combining academic papers and open-source projects",
"user_invocable": true
}
Research Survey
Conduct a comprehensive research survey combining academic and open-source sources.
Instructions
- Academic Search: Use WebSearch to find 5-10 recent papers on
${topic}from arxiv, ACL, NeurIPS, ICML, etc. - Open Source Search: Use WebSearch to find top GitHub repos related to
${topic}(sort by stars, recency) - Industry Search: Look for blog posts, technical reports from major AI labs (OpenAI, Anthropic, Google, Meta)
- Synthesize findings into a structured report:
- Executive Summary
- Academic Landscape (key papers, trends)
- Open Source Landscape (key projects, comparisons)
- Industry Trends
- Gap Analysis
- Recommendations
- Write the report to
${output}orrepo_pages/surveys/${topic}-survey.mdif no output specified - Use markdown tables for comparisons, include all source links
版本历史
- a8914bb 当前 2026-07-05 12:09


