Skills for Real Engineers. Straight from my .agents directory.
How to install research-report
install research-report
# install the Deep-Research-skills pack with the skills CLI
npx skills add Weizhena/Deep-Research-skills# or add only this skill to Claude Code: copy the folder
skills/research-zh/research-report/
# into
~/.claude/skills/research-report/ (all projects)
.claude/skills/research-report/ (one project)
1# Research Report - 汇总报告23## 触发方式4`/research-report`56## 执行流程78### Step 1: 定位结果目录9在当前工作目录查找 `*/outline.yaml`,读取topic和output_dir配置。1011### Step 2: 扫描可选摘要字段12读取所有JSON结果,提取适合在目录中显示的字段(数值型、简短指标),例如:13- github_stars14- google_scholar_cites15- swe_bench_score16- user_scale17- valuation18- release_date1920使用AskUserQuestion询问用户:21- 目录中除了item名称外,还需要显示哪些字段?22- 提供动态选项列表(基于实际JSON中存在的字段)2324### Step 3: 生成Python转换脚本25在 `{topic}/` 目录下生成 `generate_report.py`,脚本要求:26- 读取output_dir下所有JSON27- 读取fields.yaml获取字段结构28- 覆盖每个JSON的所有字段值29- 跳过值包含[不确定]的字段30- 跳过uncertain数组中列出的字段31- 生成markdown报告格式:目录(带锚点跳转+用户选择的摘要字段)+ 详细内容(按字段分类)32- 保存到 `{topic}/report.md`3334**目录格式要求**:35- 必须包含每一个item36- 每个item显示:序号、名称(锚点链接)、用户选择的摘要字段37- 示例:`1. [GitHub Copilot](#github-copilot) - Stars: 10k | Score: 85%`3839#### 脚本技术要点(必须遵循)4041**1. JSON结构兼容**42支持两种JSON结构:43- 扁平结构:字段直接在顶层 `{"name": "xxx", "release_date": "xxx"}`44- 嵌套结构:字段在category子dict中 `{"basic_info": {"name": "xxx"}, "technical_features": {...}}`4546字段查找顺序:顶层 -> category映射key -> 遍历所有嵌套dict4748**2. Category多语言映射**49fields.yaml的category名与JSON的key可能是任意组合(中中、中英、英中、英英)。必须建立双向映射:50```python51CATEGORY_MAPPING = {52 "基本信息": ["basic_info", "基本信息"],53 "技术特性": ["technical_features", "technical_characteristics", "技术特性"],54 "性能指标": ["performance_metrics", "performance", "性能指标"],55 "里程碑意义": ["milestone_significance", "milestones", "里程碑意义"],56 "商业信息": ["business_info", "commercial_info", "商业信息"],57 "竞争与生态": ["competition_ecosystem", "competition", "竞争与生态"],58 "历史沿革": ["history", "历史沿革"],59…
Excerpt shown under the repository's MIT license.
More skills in Deep-Research-skills
Keep exploring
Similar skill repositories
39 skills
An agentic skills framework & software development methodology that works.
15 skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
CC1 skill
Public repository for Agent Skills
20 skills
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
CCCU6 skills
An AI skill that provides design intelligence for building professional UI/UX across multiple platforms.
CCCXCUGH6 skills
