add-wikipedia-references

Add Wikipedia References to Concepts

This skill adds Wikipedia reference links to concept cards that are missing them. It searches for relevant Wikipedia articles and adds them to the concept's references array.

When to Use

• When you want to enrich concepts with authoritative Wikipedia references
• After bulk-importing concepts that lack references
• To improve the quality and credibility of concept cards

Prerequisites

• Concepts are stored in /home/dsebastien/wks/concept-cards/src/data/concepts/
• Each concept is a JSON file with a references array

Step-by-Step Process

Step 1: Find Concepts Without Wikipedia References

Run these commands to find concepts that don't have a Wikipedia reference:

# List concepts without Wikipedia references (using grep -L for "files without match")
grep -L '"url": "https://en.wikipedia.org' /home/dsebastien/wks/concept-cards/src/data/concepts/*.json | xargs -n1 basename | sed 's/.json$//'

# Count concepts without Wikipedia references
grep -L '"url": "https://en.wikipedia.org' /home/dsebastien/wks/concept-cards/src/data/concepts/*.json | wc -l

# Get first 20 concepts without Wikipedia (for batch processing)
grep -L '"url": "https://en.wikipedia.org' /home/dsebastien/wks/concept-cards/src/data/concepts/*.json | head -20 | xargs -n1 basename | sed 's/.json$//'

Step 2: For Each Concept, Search Wikipedia

For each concept without a Wikipedia reference:

1. Read the concept file to get the name and aliases
2. Search Wikipedia using the concept name (and aliases if needed)
3. Verify the article is relevant - the Wikipedia article should match the concept's meaning

Search Strategy

1. Primary search: Use the exact concept name

   • Example: Concept "Atomic Notes" → Search "Atomic Notes wikipedia"

2. Fallback searches (if primary fails):

   • Try aliases: "Evergreen Notes wikipedia"
   • Try broader terms: "Note-taking method wikipedia"
   • Try the concept with context: "Atomic Notes Zettelkasten wikipedia"

3. Wikipedia URL patterns:

   • Standard: https://en.wikipedia.org/wiki/Article_Name
   • With underscores: https://en.wikipedia.org/wiki/Article_Name_Here

Step 3: Verify the Wikipedia Article

Before adding a reference, verify:

1. The article exists - Navigate to the URL or fetch it
2. The article is relevant - It should describe the same concept
3. The article is specific enough - Prefer specific articles over general ones

Common Pitfalls

• Disambiguation pages: If you land on a disambiguation page, choose the most relevant article
• Redirects: Wikipedia may redirect to a differently-named article - use the final URL
• Different meanings: Some terms have multiple meanings - ensure the article matches your concept's domain (e.g., "Flow" in psychology vs. physics)

Step 4: Add the Reference

Edit the concept JSON file to add the Wikipedia reference to the references array:

{
  "references": [
    {
      "title": "Article Name - Wikipedia",
      "url": "https://en.wikipedia.org/wiki/Article_Name",
      "type": "website"
    }
  ]
}

Reference Format Rules

• title: Use format "Article Name - Wikipedia"
• url: Use the canonical Wikipedia URL (with underscores, not encoded)
• type: Always "website"

Step 5: Handle Edge Cases

No Wikipedia Article Exists

If no relevant Wikipedia article exists for a concept:

• Skip the concept
• Log it for reference
• Consider searching for related concepts that might have articles

Multiple Relevant Articles

If multiple Wikipedia articles are relevant:

• Add the most specific/relevant one
• Optionally add others as additional references

Concept is Too Niche

Some concepts may be too specific for Wikipedia:

• Personal productivity methods (e.g., specific journaling techniques)
• Proprietary frameworks
• Very recent concepts

In these cases, skip the concept.

Batch Processing

For processing multiple concepts efficiently:

Option A: Process in Batches

# Get first 20 concepts without Wikipedia references
grep -L '"url": "https://en.wikipedia.org' /home/dsebastien/wks/concept-cards/src/data/concepts/*.json | head -20 | xargs -n1 basename | sed 's/.json$//'

# Get concepts 21-40 (skip first 20)
grep -L '"url": "https://en.wikipedia.org' /home/dsebastien/wks/concept-cards/src/data/concepts/*.json | tail -n +21 | head -20 | xargs -n1 basename | sed 's/.json$//'

Then process each concept using the steps above.

Option B: Use Sub-Agents

For large batches, spawn sub-agents to process concepts in parallel:

• Each sub-agent handles 5-10 concepts
• Sub-agents search Wikipedia and update files
• Main agent coordinates and verifies

Example Workflow

Example 1: Adding Wikipedia to "pomodoro-technique"

1. Read concept:

   cat /home/dsebastien/wks/concept-cards/src/data/concepts/pomodoro-technique.json
   

2. Search Wikipedia:

   • Search: "Pomodoro Technique wikipedia"
   • Found: https://en.wikipedia.org/wiki/Pomodoro_Technique

3. Verify: The article describes the same time management technique

4. Edit the file:

   "references": [
     {
       "title": "Pomodoro Technique - Wikipedia",
       "url": "https://en.wikipedia.org/wiki/Pomodoro_Technique",
       "type": "website"
     }
   ]
   

Example 2: Concept with No Wikipedia Article

1. Concept: "progressive-summarization"
2. Search: "Progressive Summarization wikipedia" → No results
3. Fallback: "Tiago Forte Progressive Summarization" → No Wikipedia article
4. Decision: Skip - this is a proprietary method without a Wikipedia entry

Verification

After adding references, verify the build succeeds:

npm run build 2>&1 | tail -10

Tips for Efficiency

1. Process alphabetically to track progress easily
2. Use WebSearch to find Wikipedia URLs quickly
3. Batch similar concepts (e.g., all psychology concepts together)
4. Skip obvious misses - proprietary methods, very recent concepts
5. Use Playwright for disambiguation pages or to verify article content

Reference Quality Guidelines

• Prefer English Wikipedia (en.wikipedia.org)
• Use canonical URLs (not mobile, not shortened)
• Verify article stability - well-established articles are better than stubs
• Match the concept's domain - "Flow" should link to Flow (psychology), not Flow (physics)