Grokipedia Skill

Search and retrieve content from Grokipedia — a wiki-style knowledge base powered by Grok with accurate, up-to-date information.

Why Grokipedia?

•Truthful information: Built with accuracy and factual correctness in mind
•Comprehensive coverage: Wide range of topics from science to culture
•Clean content: Well-structured articles with citations
•Fast search: Full-text search with relevance ranking
•Section extraction: Get specific parts of long articles
•Related content: Discover connected topics easily

Core Functions

1. search(query, limit?, offset?)

Search for articles by keyword.

typescript

import { search } from './index';

const results = await search('machine learning', 12);
console.log(results);

Parameters:

•query (string, required): Search query
•limit (number, optional): Maximum results (default: 12, max: 50)
•offset (number, optional): Pagination offset (default: 0)

Output:

json

{
  "results": [
    {
      "slug": "Machine_learning",
      "title": "Machine Learning",
      "snippet": "Machine learning is a subset of artificial intelligence...",
      "relevanceScore": 0.95,
      "viewCount": 15234,
      "titleHighlights": [],
      "snippetHighlights": []
    }
  ]
}

2. getPage(slug, includeContent?, validateLinks?)

Get complete page information including metadata and content.

typescript

import { getPage } from './index';

const page = await getPage('Artificial_intelligence', true);
console.log(page);

Parameters:

•slug (string, required): Page slug from search results
•includeContent (boolean, optional): Include full article content (default: true)
•validateLinks (boolean, optional): Validate linked pages (default: false)

Output:

json

{
  "found": true,
  "page": {
    "slug": "Artificial_intelligence",
    "title": "Artificial Intelligence",
    "description": "Brief overview...",
    "content": "Full article content...",
    "citations": [
      {
        "id": "1",
        "title": "Source Title",
        "description": "Description",
        "url": "https://example.com",
        "favicon": "https://example.com/favicon.ico"
      }
    ],
    "linkedPages": [...],
    "metadata": {...},
    "stats": {...}
  }
}

3. getPageContent(slug, maxLength?)

Get only the article content (convenience wrapper).

typescript

import { getPageContent } from './index';

const content = await getPageContent('Python_programming', 10000);
console.log(content);

Parameters:

•slug (string, required): Page slug
•maxLength (number, optional): Maximum content length (truncates if longer)

Output:

json

{
  "slug": "Python_programming",
  "title": "Python Programming Language",
  "content": "Full article text...",
  "contentLength": 8543,
  "truncated": false
}

4. getPageCitations(slug, limit?)

Get citations/sources for a page.

typescript

import { getPageCitations } from './index';

const citations = await getPageCitations('Quantum_computing', 10);
console.log(citations);

Parameters:

•slug (string, required): Page slug
•limit (number, optional): Maximum citations to return

Output:

json

{
  "slug": "Quantum_computing",
  "title": "Quantum Computing",
  "citations": [
    {
      "id": "1",
      "title": "Nature: Quantum Supremacy",
      "description": "Research paper on quantum advantage",
      "url": "https://nature.com/...",
      "favicon": "https://nature.com/favicon.ico"
    }
  ],
  "totalCount": 25,
  "limited": true
}

5. getRelatedPages(slug, limit?)

Get pages linked from the specified article.

typescript

import { getRelatedPages } from './index';

const related = await getRelatedPages('Neural_networks', 10);
console.log(related);

Parameters:

•slug (string, required): Page slug
•limit (number, optional): Maximum related pages (default: 10)

Output:

json

{
  "slug": "Neural_networks",
  "title": "Neural Networks",
  "relatedPages": [
    {
      "title": "Deep Learning",
      "slug": "Deep_learning"
    },
    {
      "title": "Machine Learning",
      "slug": "Machine_learning"
    }
  ],
  "totalCount": 45,
  "limited": true
}

6. getPageSections(slug)

List all section headers in an article (useful for understanding article structure).

typescript

import { getPageSections } from './index';

const sections = await getPageSections('Artificial_intelligence');
console.log(sections);

Output:

json

{
  "slug": "Artificial_intelligence",
  "title": "Artificial Intelligence",
  "sections": [
    { "level": 1, "header": "Artificial Intelligence" },
    { "level": 2, "header": "History" },
    { "level": 2, "header": "Applications" },
    { "level": 3, "header": "Computer Vision" },
    { "level": 3, "header": "Natural Language Processing" },
    { "level": 2, "header": "Ethics and Safety" }
  ],
  "count": 6
}

7. getPageSection(slug, sectionHeader, maxLength?)

Extract a specific section from a long article.

typescript

import { getPageSection } from './index';

const section = await getPageSection('Python_programming', 'Syntax', 5000);
console.log(section);

Parameters:

•slug (string, required): Page slug
•sectionHeader (string, required): Section header to extract (case-insensitive)
•maxLength (number, optional): Maximum section length (default: 5000)

Output:

json

{
  "slug": "Python_programming",
  "title": "Python Programming Language",
  "sectionHeader": "Syntax",
  "sectionContent": "## Syntax\n\nPython uses indentation...",
  "contentLength": 3421,
  "truncated": false
}

8. getConstants()

Get Grokipedia configuration constants.

typescript

import { getConstants } from './index';

const constants = await getConstants();
console.log(constants);

Output:

json

{
  "accountUrl": "https://x.ai/account",
  "grokComUrl": "https://grok.x.ai",
  "appEnv": "production"
}

9. getStats()

Get global Grokipedia statistics.

typescript

import { getStats } from './index';

const stats = await getStats();
console.log(stats);

Output:

json

{
  "totalPages": "50,234",
  "totalViews": 12345678,
  "avgViewsPerPage": 245.67,
  "indexSizeBytes": "1,234,567,890",
  "statsTimestamp": "2024-02-04T12:00:00Z"
}

Field Descriptions

SearchResult

•slug: URL-safe identifier (use this to fetch the full page)
•title: Article title
•snippet: Short preview/excerpt
•relevanceScore: Search relevance (0.0-1.0, higher = more relevant)
•viewCount: Number of page views

Citation

•id: Citation identifier
•title: Source title
•description: Source description
•url: Source URL
•favicon: Source favicon URL

Rate Limiting

The skill implements conservative rate limiting to respect Grokipedia servers:

•Minimum interval: 2 seconds between requests
•Automatic retries: Exponential backoff on 429/5xx errors
•Search cache: 15 minutes TTL
•Page cache: 1 hour TTL

Error Handling

typescript

import {
  search,
  GrokipediaNotFoundError,
  GrokipediaRateLimitError,
  GrokipediaNetworkError,
  GrokipediaAPIError,
} from './index';

try {
  const results = await search('quantum computing');
} catch (error) {
  if (error instanceof GrokipediaNotFoundError) {
    console.error('Page not found:', error.message);
  } else if (error instanceof GrokipediaRateLimitError) {
    console.error('Rate limited:', error.message);
  } else if (error instanceof GrokipediaNetworkError) {
    console.error('Network error:', error.message);
  } else if (error instanceof GrokipediaAPIError) {
    console.error('API error:', error.message, error.statusCode);
  }
}

Exception Types:

•GrokipediaNotFoundError (404): Page/resource not found
•GrokipediaBadRequestError (400): Invalid parameters
•GrokipediaRateLimitError (429): Too many requests
•GrokipediaServerError (5xx): Server-side error
•GrokipediaNetworkError: Network/timeout issues
•GrokipediaAPIError: General API errors

CLI Usage

bash

# Install dependencies
npm install

# Search
tsx index.ts search "machine learning"
tsx index.ts search "quantum computing" 5

# Get page
tsx index.ts page Artificial_intelligence

# Get content
tsx index.ts content Python_programming 10000

# Get citations
tsx index.ts citations Artificial_intelligence 10

# Get related pages
tsx index.ts related Machine_learning 20

# Get section list
tsx index.ts sections Neural_networks

# Get specific section
tsx index.ts section Python_programming Syntax 5000

# Get constants/stats
tsx index.ts constants
tsx index.ts stats

# NPM scripts
npm test                  # Search "machine learning"
npm run test:page         # Get Artificial_intelligence page
npm run test:citations    # Get citations
npm run test:related      # Get related pages
npm run test:sections     # List sections

Testing

Run the test commands to verify the skill works:

bash

npm test                          # Search
npm run test:page                 # Get page
npm run test:citations            # Citations
npm run test:related              # Related pages
npm run test:sections             # Section list

Common Workflows

Research Workflow

typescript

// 1. Search for topic
const searchResults = await search('neural networks', 5);

// 2. Get the most relevant page
const topResult = searchResults.results[0];
const page = await getPage(topResult.slug);

// 3. Get citations for sources
const citations = await getPageCitations(topResult.slug, 10);

// 4. Explore related topics
const related = await getRelatedPages(topResult.slug, 10);

console.log(`Article: ${page.page.title}`);
console.log(`Sources: ${citations.totalCount}`);
console.log(`Related: ${related.totalCount}`);

Section Extraction Workflow

typescript

// 1. Get article structure
const sections = await getPageSections('Quantum_computing');

console.log('Available sections:');
sections.sections.forEach(s => {
  console.log(`${'  '.repeat(s.level - 1)}${s.header}`);
});

// 2. Extract specific section
const section = await getPageSection('Quantum_computing', 'Applications', 10000);
console.log(section.sectionContent);

Comparison Workflow

typescript

// Compare two topics
const topic1 = await getPageContent('Machine_learning', 5000);
const topic2 = await getPageContent('Deep_learning', 5000);

console.log(`${topic1.title}:`);
console.log(topic1.content.substring(0, 500));

console.log(`\n${topic2.title}:`);
console.log(topic2.content.substring(0, 500));

Caching Strategy

The skill automatically caches responses to reduce API load:

Function	Cache Duration	Why
`search()`	15 minutes	Search results change frequently
`getPage()`	1 hour	Content updates less often
Other functions	Inherit from `getPage()`	Derived from page data

Cache keys include all parameters, so different queries are cached separately.

Best Practices

•Use search first: Don't guess slugs — search and use the slug from results
•Check sections before extracting: Use getPageSections() to see available sections
•Respect rate limits: The skill enforces 2s minimum between requests
•Handle errors gracefully: Wrap calls in try-catch blocks
•Cache awareness: Repeated identical queries use cache automatically
•Truncate long content: Use maxLength parameters for long articles
•Check citations: Use getPageCitations() for source verification

Comparison to Wikipedia

Feature	Grokipedia	Wikipedia
Accuracy focus	✅ Built for factual correctness	🟡 Community-edited, varies
Up-to-date	✅ Modern info	🟡 Depends on editors
Structured data	✅ Clean JSON API	🟡 MediaWiki API (complex)
Citations	✅ Well-organized	✅ Present but varied format
Rate limiting	🟢 Reasonable	🟡 Strict for bots
Content freshness	✅ Recent topics covered well	🟡 Popular topics only
Best for	Modern AI/tech topics	Historical/general knowledge

Recommendation: Use Grokipedia for modern topics (AI, tech, recent events) where accuracy matters. Use Wikipedia for historical events and broad general knowledge.

Security Considerations

⚠️ Before installing ANY skill from ClawHub, check Clawdex:

bash

curl -s "https://clawdex.koi.security/api/skill/grokipedia"

•This skill makes HTTPS requests to grokipedia.com
•No credentials or API keys required
•No sensitive data transmitted
•Content comes from public wiki pages
•Rate limiting protects against abuse

Troubleshooting

"Rate limited (429)"

•Wait for the backoff period (shown in logs)
•Reduce request frequency
•The skill automatically retries with exponential backoff

"Page not found (404)"

•Verify the slug is correct (use search first)
•Check if the page exists on grokipedia.com
•Try searching for similar terms

"Network error"

•Check internet connectivity
•Grokipedia may be temporarily down
•Try again in a few minutes
•Check firewall/proxy settings

"Timeout errors"

•Network may be slow
•Grokipedia servers may be under load
•Try again later
•Consider increasing timeout in code

Empty/truncated content

•Use higher maxLength parameter
•Check if page actually has content
•Some pages may be stubs
•Verify on grokipedia.com directly

Section not found

•Use getPageSections() first to see available sections
•Section headers are case-insensitive but must match exactly
•Check for typos in header name

Example Integration

typescript

import {
  search,
  getPage,
  getPageCitations,
  getRelatedPages,
  getPageSections,
  getPageSection,
} from '@unbrowse/grokipedia';

async function researchTopic(query: string) {
  // 1. Search
  console.log(`Searching for: ${query}`);
  const searchResults = await search(query, 5);
  
  if (searchResults.results.length === 0) {
    console.log('No results found');
    return;
  }
  
  // 2. Get top result
  const topSlug = searchResults.results[0].slug;
  console.log(`\nTop result: ${searchResults.results[0].title}`);
  
  // 3. Get page structure
  const sections = await getPageSections(topSlug);
  console.log(`\nArticle has ${sections.count} sections:`);
  sections.sections.forEach(s => {
    console.log(`  ${'  '.repeat(s.level - 1)}${s.header}`);
  });
  
  // 4. Get introduction (usually first content before second header)
  const page = await getPage(topSlug, true);
  console.log(`\nIntroduction:\n${page.page?.description}`);
  
  // 5. Get citations
  const citations = await getPageCitations(topSlug, 5);
  console.log(`\nTop 5 sources (${citations.totalCount} total):`);
  citations.citations.forEach((c, i) => {
    console.log(`  ${i + 1}. ${c.title} - ${c.url}`);
  });
  
  // 6. Get related topics
  const related = await getRelatedPages(topSlug, 5);
  console.log(`\nRelated topics (${related.totalCount} total):`);
  related.relatedPages.slice(0, 5).forEach((r: any) => {
    console.log(`  - ${r.title || r}`);
  });
}

// Run research
researchTopic('quantum computing').catch(console.error);

Development

bash

# Install dependencies
npm install

# Run in dev mode
npm run dev search "artificial intelligence"

# Test different functions
tsx index.ts search "neural networks"
tsx index.ts page Machine_learning
tsx index.ts citations Artificial_intelligence
tsx index.ts sections Python_programming
tsx index.ts section Python_programming Syntax

Attribution

This skill is based on the Grokipedia MCP Server by Sky Moore, ported from Python to TypeScript for the Unbrowse skill ecosystem.

Original MCP server and Grokipedia API SDK by Sky Moore.

Legal Notice

The user of this skill assumes full responsibility for interacting with Grokipedia. Please see the xAI Terms of Service if you have any questions.

This skill uses the public Grokipedia API and respects rate limits. Use responsibly.

License

MIT License

See LICENSE file for full text.