De Wikipedia Org

Wikipedia German (de.wikipedia.org) Access Skill

Extract structured tabular data from German Wikipedia pages, including award nominations/winners, navigation tables, and general wikitable data.

Overview

This skill provides reliable access to structured data tables on German Wikipedia (de.wikipedia.org). It uses Playwright for browser automation to bypass Wikipedia's anti-bot protection and BeautifulSoup for robust HTML parsing.

Functions

search

Fetch a Wikipedia page and return metadata.

Parameters:

• url (string, optional): Full Wikipedia URL
• page_title (string, optional): Wikipedia page title (will construct URL)

Returns:

{
  "success": true,
  "url": "https://de.wikipedia.org/wiki/...",
  "title": "Page Title – Wikipedia",
  "table_count": 2,
  "content_length": 191048
}

extract_tables

Extract all wikitable data from a Wikipedia page.

Parameters:

• url (string, required): Full Wikipedia URL

Returns:

{
  "success": true,
  "url": "...",
  "title": "...",
  "total_tables": 2,
  "tables": [
    {
      "table_index": 0,
      "headers": ["Jahr", "Interpret", "Nationalität", "Werk", "Weitere nominierte Künstler"],
      "row_count": 21,
      "column_count": 5,
      "data": [
        {"row_index": 1, "cells": ["1991|20. Februar 1991", "Big Daddy Kane, Ice-T, ...", ...]}
      ]
    }
  ]
}

extract_awards

Extract award nominations/winners data with specialized parsing.

Parameters:

• url (string, required): Full Wikipedia URL to the awards page

Returns:

{
  "success": true,
  "url": "...",
  "title": "...",
  "headers": ["Jahr", "Interpret", "Nationalität", "Werk", "Weitere nominierte Künstler"],
  "total_entries": 21,
  "awards": [
    {
      "year": "1991",
      "year_full": "1991|20. Februar 1991",
      "artist": "Big Daddy Kane, Ice-T, Kool Moe Dee, ...",
      "nationality": "Vereinigte Staaten",
      "work": "Back on the Block",
      "other_nominees": ["Digital Underground – The Humpty Dance", ...],
      "raw_cells": [...]
    }
  ]
}

fetch_page

Fetch page and return raw HTML content.

Parameters:

• url (string, required): Full Wikipedia URL

Returns:

{
  "success": true,
  "url": "...",
  "title": "...",
  "html": "<div>...</div>",
  "content_length": 191048
}

Usage Examples

Extract Grammy Awards Data

result = await execute({
    "function": "extract_awards",
    "url": "https://de.wikipedia.org/wiki/Grammy_Award_for_Best_Rap_Performance_by_a_Duo_or_Group"
})

for award in result["awards"]:
    print(f"{award['year']}: {award['artist']} - {award['work']}")

Search by Page Title

result = await execute({
    "function": "search",
    "page_title": "Grammy Awards"
})

Extract All Tables

result = await execute({
    "function": "extract_tables",
    "url": "https://de.wikipedia.org/wiki/Liste_der_Grammy-Gewinner"
})

for table in result["tables"]:
    print(f"Table {table['table_index']}: {table['row_count']} rows")

Data Format Notes

Awards Table Structure

German Wikipedia awards tables typically follow this structure:

• Jahr: Year (may include ceremony date)
• Interpret: Artist/performer name(s)
• Nationalität: Country/nationality
• Werk: Work/title
• Weitere nominierte Künstler: Other nominees

Text Separators

Cells with multiple lines use pipe (|) characters as separators in the extracted text to preserve structure while allowing easy parsing:

"1991|20. Februar 1991"  # Year and date
"Big Daddy Kane|,|Ice-T|,|..."  # Multiple artists

Implementation Notes

Browser Automation

This skill uses Playwright with headless Chromium because:

1. Wikipedia blocks direct HTTP requests from scripts (returns 403)
2. User-Agent headers alone are not sufficient
3. Browser context provides proper request signatures

Rate Limiting

Consider implementing rate limiting for bulk requests to avoid being blocked by Wikipedia's servers.

Character Encoding

German Wikipedia uses UTF-8 encoding. The skill handles German characters (ä, ö, ü, ß) and special formatting correctly.

Error Handling

All errors are returned in the response with explicit error types:

{
  "success": false,
  "error": "Page load timeout",
  "error_type": "timeout"
}

Common error types:

• timeout: Page load exceeded timeout
• no_table: No wikitable found in content
• index_error: Table index out of range
• missing_parameter: Required parameter not provided
• invalid_function: Unknown function name