List Knowledge Bases

Retrieve all available knowledge bases in your system.

Overview

The Knowledge Bases endpoint provides a comprehensive list of all registered knowledge bases, including their metadata and creation details. This is essential for understanding what data sources are available for querying and analysis.

info

Knowledge bases are automatically created when you register Google Sheets or other data sources. Each knowledge base represents a searchable, indexed version of your data.

Endpoint

GET /knowledge-bases

Request

This endpoint requires no parameters or request body.

GET /knowledge-bases

Response

Success Response

{
  "status": "success",
  "knowledge_bases": [
    {
      "name": "sales_data_kb",
      "engine": "chromadb",
      "created_at": "2025-06-17 10:00:00"
    },
    {
      "name": "customer_feedback_kb",
      "engine": "chromadb", 
      "created_at": "2025-06-18 14:30:00"
    },
    {
      "name": "product_inventory_kb",
      "engine": "chromadb",
      "created_at": "2025-06-19 09:15:00"
    }
  ]
}

Error Response

{
  "status": "error",
  "message": "Unable to retrieve knowledge bases"
}

Response Fields

Success Response Fields

Field	Type	Description
status	string	Always "success" for successful requests
knowledge_bases	array	Array of knowledge base objects

Knowledge Base Object Fields

Field	Type	Description
name	string	Unique identifier for the knowledge base
engine	string	Vector database engine used (typically "chromadb")
created_at	string	When the knowledge base was created (YYYY-MM-DD HH:MM:SS format)

Common Error Messages

Error Message	Cause	Solution
"Unable to retrieve knowledge bases"	System or database issue	Try again later or contact support
"Database connection error"	Connection problem	Check system status or try again
"Access denied"	Insufficient permissions	Verify your account permissions

Use Cases

📊 System Overview

GET /knowledge-bases

Get a complete overview of all available data sources for analysis.

🔍 Data Source Discovery

Before querying agents, check what knowledge bases are available for different topics.

📈 Usage Monitoring

Track when knowledge bases were created to understand data ingestion patterns.

🎯 Integration Planning

Identify existing knowledge bases before registering new ones to avoid duplicates.

Knowledge Base Naming Conventions

Knowledge bases are typically named based on:

• Source Data: sales_data_kb, customer_orders_kb
• Department: hr_policies_kb, marketing_campaigns_kb
• Time Period: q1_2024_sales_kb, annual_reports_kb
• Data Type: feedback_kb, inventory_kb, transactions_kb

Integration Examples

Check Available Knowledge Bases

// Fetch all knowledge bases
async function getKnowledgeBases() {
  try {
    const response = await fetch('/knowledge-bases');
    const data = await response.json();
    
    if (data.status === 'success') {
      console.log('Available knowledge bases:');
      data.knowledge_bases.forEach(kb => {
        console.log(`- ${kb.name} (created: ${kb.created_at})`);
      });
      return data.knowledge_bases;
    } else {
      console.error('Failed to fetch knowledge bases:', data.message);
      return [];
    }
  } catch (error) {
    console.error('Error fetching knowledge bases:', error);
    return [];
  }
}

Filter by Creation Date

// Get recently created knowledge bases
function getRecentKnowledgeBases(knowledgeBases, days = 7) {
  const cutoffDate = new Date();
  cutoffDate.setDate(cutoffDate.getDate() - days);
  
  return knowledgeBases.filter(kb => {
    const createdDate = new Date(kb.created_at);
    return createdDate >= cutoffDate;
  });
}

Build Knowledge Base Selector

// Create dropdown options for knowledge base selection
function createKBSelector(knowledgeBases) {
  const selector = document.getElementById('kb-selector');
  
  knowledgeBases.forEach(kb => {
    const option = document.createElement('option');
    option.value = kb.name;
    option.textContent = `${kb.name} (${kb.created_at})`;
    selector.appendChild(option);
  });
}

Knowledge Base Lifecycle

1. Creation

Knowledge bases are created when you register data sources:

POST /register-sheet
{
  "id": "sales-data-2024",
  "spreadsheet_id": "...",
  "sheet_name": "Sales"
}

2. Indexing

Data is processed and indexed using ChromaDB for semantic search.

3. Querying

Knowledge bases become available for agent queries and semantic search.

4. Updates

Re-registering with the same ID updates the knowledge base content.

Best Practices

Regular Monitoring

• Check Regularly: Monitor available knowledge bases to understand your data landscape
• Track Creation Dates: Identify when new data sources were added
• Verify Completeness: Ensure all expected knowledge bases are present

Naming Strategy

• Use Descriptive Names: Make knowledge base purposes clear
• Include Dates: Add time periods for time-sensitive data
• Consistent Conventions: Follow team naming standards

Integration Patterns

• Cache Results: Store knowledge base list to avoid repeated API calls
• Dynamic UI: Build interfaces that adapt to available knowledge bases
• Error Handling: Gracefully handle cases where knowledge bases are unavailable

Related Endpoints

After getting your knowledge bases list, you can:

1. Query Specific Knowledge Bases: Use the semantic search endpoint
2. Create AI Agents: Reference knowledge bases when setting up agents
3. Register New Data: Add more knowledge bases via sheet registration

System Information

tip

Vector Engine: All knowledge bases use ChromaDB for high-performance semantic search and similarity matching.

Automatic Updates: Knowledge bases are automatically updated when you re-register data sources with the same ID.

Next Steps

Once you have your knowledge bases list:

1. Semantic Search: Query specific knowledge bases for relevant information
2. Agent Creation: Use knowledge bases to power your AI agents
3. Data Analysis: Understand what data is available for analysis
4. System Maintenance: Monitor and manage your knowledge base collection