Agent Skills
› mhattingpete/claude-skills-marketplace
› file-operations
file-operations
GitHub提供文件元数据查询、统计分析(如行数、大小)及内容模式搜索能力,支持代码质量评估与文件对比,全程只读不修改。
Trigger Scenarios
用户请求获取文件信息或统计
需要分析代码结构或查找特定模式
Install
npx skills add mhattingpete/claude-skills-marketplace --skill file-operations -g -y
SKILL.md
Frontmatter
{
"name": "file-operations",
"description": "Analyze files and get detailed metadata including size, line counts, modification times, and content statistics. Use when users request file information, statistics, or analysis without modifying files."
}
File Operations
Analyze files and retrieve metadata using Claude's native tools without modifying files.
When to Use
- "analyze [file]"
- "get file info for [file]"
- "how many lines in [file]"
- "compare [file1] and [file2]"
- "file statistics"
Core Operations
File Size & Metadata
stat -f "%z bytes, modified %Sm" [file_path] # Single file
ls -lh [directory] # Multiple files
du -h [file_path] # Human-readable size
Line Counts
wc -l [file_path] # Single file
wc -l [file1] [file2] # Multiple files
find [dir] -name "*.py" | xargs wc -l # Directory total
Content Analysis
Use Read to analyze structure, then count functions/classes/imports.
Pattern Search
Grep(pattern="^def ", output_mode="count", path="src/") # Count functions
Grep(pattern="TODO|FIXME", output_mode="content", -n=true) # Find TODOs
Grep(pattern="^import ", output_mode="count") # Count imports
Find Files
Glob(pattern="**/*.py")
Workflow Examples
Comprehensive File Analysis
- Get size/mod time:
stat -f "%z bytes, modified %Sm" file.py - Count lines:
wc -l file.py - Read file:
Read(file_path="file.py") - Count functions:
Grep(pattern="^def ", output_mode="count") - Count classes:
Grep(pattern="^class ", output_mode="count")
Compare File Sizes
- Find files:
Glob(pattern="src/**/*.py") - Get sizes:
ls -lh src/**/*.py - Total size:
du -sh src/*.py
Code Quality Metrics
- Total lines:
find . -name "*.py" | xargs wc -l - Test files:
find . -name "test_*.py" | wc -l - TODOs:
Grep(pattern="TODO|FIXME|HACK", output_mode="count")
Find Largest Files
find . -type f -not -path "./node_modules/*" -exec du -h {} + | sort -rh | head -20
Best Practices
- Non-destructive: Use Read/stat/wc, never modify
- Efficient: Read small files fully, use Grep for large files
- Context-aware: Compare to project averages, suggest optimizations
Integration
Works with:
- code-auditor: Comprehensive analysis
- code-transfer: After identifying large files
- codebase-documenter: Understanding file purposes
Version History
- b5b34bc Current 2026-08-20 08:47


