Claude-skill-registry add-wikipedia-references

Add Wikipedia reference links to concepts that don't have one. Searches for relevant Wikipedia articles and adds them to the references array.

install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/add-wikipedia-references" ~/.claude/skills/majiayu000-claude-skill-registry-add-wikipedia-references && rm -rf "$T"
manifest: skills/data/add-wikipedia-references/SKILL.md
source content

Add Wikipedia References

Add Wikipedia links to concepts missing them.

Find Concepts Without Wikipedia

# List concepts without Wikipedia
grep -L '"url": "https://en.wikipedia.org' src/data/concepts/*.json | xargs -n1 basename | sed 's/.json$//'

# Count
grep -L '"url": "https://en.wikipedia.org' src/data/concepts/*.json | wc -l

# First 20 for batch processing
grep -L '"url": "https://en.wikipedia.org' src/data/concepts/*.json | head -20 | xargs -n1 basename | sed 's/.json$//'

Workflow Per Concept

  1. Read concept - get name, aliases
  2. Search Wikipedia - use concept name, then aliases if needed
  3. Verify relevance - article must match concept's meaning/domain
  4. Add reference:
{
  "references": [
    {
      "title": "Article Name - Wikipedia",
      "url": "https://en.wikipedia.org/wiki/Article_Name",
      "type": "website"
    }
  ]
}

Search Strategy

  1. Primary: exact concept name + "wikipedia"
  2. Fallback: aliases, broader terms, concept with context
  3. Handle disambiguation pages: choose most relevant article
  4. Use final URL after redirects

Skip When

  • No relevant Wikipedia article exists
  • Concept too niche (proprietary methods, very recent concepts)
  • Article doesn't match concept's domain (e.g., "Flow" in wrong field)

Reference Format

  • title:
    "Article Name - Wikipedia"
  • url: canonical URL with underscores (
    https://en.wikipedia.org/wiki/Article_Name
    )
  • type:
    "website"

Batch Processing

# Batch 1-20
grep -L '"url": "https://en.wikipedia.org' src/data/concepts/*.json | head -20

# Batch 21-40
grep -L '"url": "https://en.wikipedia.org' src/data/concepts/*.json | tail -n +21 | head -20

For large batches: spawn sub-agents (5-10 concepts each).

Verify

npm run build 2>&1 | tail -10