AgentSkillsCN

url-parameter-parser

解析 CSV 文件中的 URL，并将查询参数提取为新列。在处理包含 URL 且需提取与分析参数的 CSV 文件时使用。

SKILL.md
--- frontmatter
name: url-parameter-parser
description: Parse URLs in CSV files and extract query parameters as new columns. Use when working with CSV files containing URLs that need parameter extraction and analysis.
allowed-tools: Read, Write, Bash, Glob

URL Parameter Parser

This skill extracts query parameters from URLs in CSV files and adds them as new columns.

Instructions

  1. •Identify CSV files: Look for CSV files in the current directory or specified path
  2. •Analyze URL column: Find the column containing URLs (looks for 'url' or 'URL' headers)
  3. •Extract parameters: Parse all query parameters from URLs
  4. •Create new columns: Add parameter names as new column headers
  5. •Process data: Fill new columns with parameter values (multiple values joined with '|')
  6. •Save results: Update the CSV file with new columns

Implementation

The skill uses Python to process CSV files. Run the processing script:

bash
python scripts/process_csv.py [file1.csv file2.csv ...]

If no files are specified, it processes all CSV files in the current directory.

The script will:

  • •Read CSV files with headers
  • •Extract query parameters from URLs using urllib.parse
  • •Handle multiple values for the same parameter (joined with '|')
  • •Preserve original data while adding new parameter columns
  • •Handle malformed URLs gracefully

For detailed examples, see EXAMPLES.md.

Usage Examples

Process all CSV files in current directory:

code
Parse URLs in my CSV files

Process specific CSV file:

code
Extract URL parameters from data.csv

Analyze URL parameters:

code
Show me what parameters are in the URLs from this CSV

Requirements

  • •Python 3.6 or higher with standard libraries (csv, urllib.parse)
  • •CSV files must have headers
  • •URL column should be named 'url' or 'URL'

Output

  • •Original CSV file updated with new parameter columns
  • •Multiple parameter values separated by '|'
  • •Preserves all original data
  • •Handles empty/missing parameters gracefully