Honey Jar User Guide

Overview

Honey Jars are STING’s intelligent knowledge containers that store, organize, and make your documents searchable through AI-powered semantic search. Think of them as secure, smart filing cabinets that understand the meaning of your content.

Key Features

🍯 Document Management

  • Multi-format Support: Upload PDF, Word, Markdown, JSON, HTML, and text files.
  • Bulk Upload: Drag and drop multiple files at once.
  • Real-time Processing: Watch as documents are processed and indexed.
  • Metadata Tagging: Organize documents with custom tags and categories.

🔍 Intelligent Search

  • Semantic Search: Find documents by meaning, not just keywords.
  • Vector Embeddings: Documents are converted to AI-understandable formats.
  • Relevance Scoring: Results ranked by semantic similarity.

🐝 Query with Bee Integration

  • Context-Aware Chat: Ask Bee questions about specific honey jar contents.
  • Automatic Context: Bee understands which honey jar you’re discussing.
  • Natural Language: Ask questions in plain English.

📦 Export & Sharing

  • HJX Format: STING’s proprietary Honey Jar Export format (recommended)
    • Includes all documents and metadata
    • Preserves embeddings and search capabilities
    • Can be imported into other STING instances
  • JSON Export: Plain JSON with all metadata for integration.
  • TAR Archive: Simple archive of all documents for backup.

Getting Started

Creating Your First Honey Jar

  1. Navigate to the Honey Jars tab in your dashboard
  2. Click Create Honey Jar button
  3. Fill in:
    • Name: A descriptive name for your knowledge base.
    • Description: What this honey jar contains.
    • Type: Choose visibility level:
      • Public: Accessible to all users
      • Private: Only you can access
      • Team: Shared with your team
      • Restricted: Specific user permissions.

Uploading Documents

  1. Open a honey jar by clicking on it
  2. Click the green Upload Documents button
  3. Select files or drag & drop them
  4. Wait for processing to complete (progress shown in real-time)

Note on Document Approval:

  • Admin users: Documents are uploaded immediately.
  • Honey jar owners: Documents are uploaded immediately to their own honey jars.
  • Regular users on public honey jars: Documents go to a pending queue for admin approval.
  • You’ll see a message indicating if your documents require approval.

Querying with Bee

  1. In the honey jar details view, click Query with Bee
  2. You’ll be taken to the chat interface with:
    • The honey jar context pre-loaded
    • A suggested initial question
    • Visual indicator showing which honey jar is active.
  3. Ask questions naturally - Bee will search only within that honey jar
  4. Click the X button next to the honey jar name to clear context

Exporting Honey Jars

  1. Open the honey jar you want to export
  2. Click the Export button
  3. Choose your format:
    • HJX Format (recommended): Complete export with all data
    • JSON Format: For developers and integrations.
    • TAR Archive: Simple document backup.
  4. The download will start automatically

Advanced Features

Sample Documents

New honey jars come pre-loaded with sample STING documentation:

  • Platform Overview
  • Honeypot Setup Guide
  • API Reference
  • Security Best Practices
  • Threat Analysis Patterns.

These help you understand the system and can be deleted if not needed.

Search Capabilities

The knowledge service uses advanced vector search technology:

  • Documents are chunked into semantic segments
  • Each segment is converted to a high-dimensional vector
  • Searches find conceptually similar content, not just keyword matches.

Integration with Bee

When you query with Bee while a honey jar is active:

  • Bee searches only within that specific honey jar
  • Responses are enhanced with relevant document snippets
  • Source documents are referenced in responses
  • Context remains active until manually cleared.

Best Practices

Document Organization

  1. Use Descriptive Names: Name honey jars clearly (e.g., “Q4 2024 Financial Reports”)
  2. Tag Consistently: Use standardized tags across your organization
  3. Regular Updates: Keep documents current by removing outdated versions
  4. Size Limits: Keep individual documents under 50MB for optimal performance

Security Considerations

  1. Access Control: Set appropriate visibility levels for sensitive data
  2. Regular Audits: Review who has access to your honey jars
  3. Export Carefully: Exported honey jars contain all document content
  4. Delete Securely: Removing documents permanently deletes them

Performance Tips

  1. Batch Uploads: Upload multiple related documents together
  2. Wait for Processing: Let documents fully process before searching
  3. Use Specific Queries: More specific questions yield better results
  4. Monitor Stats: Check document and embedding counts regularly

Troubleshooting

Common Issues

“Using offline data” warning

  • This appears when the knowledge service is temporarily unavailable
  • Your data is safe - try refreshing the page
  • If persistent, contact your administrator.

Upload failures

  • Check file size (max 50MB per file)
  • Ensure file format is supported
  • Verify you have upload permissions for the honey jar
  • If you see “permission denied”, you may need admin approval for uploads.

Query with Bee not working

  • Ensure you’re logged in
  • Check that the honey jar has processed documents
  • Try clearing browser cache if navigation fails.

Export taking too long

  • Large honey jars may take time to package
  • Check your browser’s download folder
  • Try a different export format if one fails.

Getting Help

For additional support:

  • Check the platform documentation in the sample honey jar
  • Contact your system administrator
  • Submit a support ticket through the help menu.

Glossary

  • Honey Jar: A knowledge container storing related documents.
  • Embeddings: Mathematical representations of document meaning.
  • Vector Search: Finding documents by conceptual similarity.
  • HJX Format: Honey Jar Export - STING’s native export format.
  • Semantic Search: Search by meaning rather than exact keywords.
  • Nectar Processing: The system that extracts and indexes document content.

Last updated: