Overview and Statistics

The Knowledge Base (KB) serves as the centralized intelligence repository for the Antimanual plugin, acting as the primary data source for all AI-driven operations. By indexing and vectorizing your existing content, the system enables your AI assistant to generate context-aware responses and high-quality content that aligns perfectly with your brand’s unique voice and established data.

• The Statistical Overview Dashboard
• Diverse Data Source Integration
• Resource and Limit Management
• Frequently Asked Questions

The Statistical Overview Dashboard

Upon accessing the Knowledge Base, users are greeted with a comprehensive Statistics Panel located at the top of the interface. This real-time dashboard provides a snapshot of your current data architecture and indexing status, allowing for efficient management of your AI’s training material. The dashboard tracks several key metrics:

• Total KB Items: An aggregate count of every individual content piece, whether it be a post, a PDF, or a URL, currently stored in the vector database.
• WordPress Posts: Specifically monitors the volume of native site content (Posts, Pages, or custom post types) that has been synced to the AI.
• Website URL: Tracks the number of external web pages that have been crawled and indexed via the scraping engine.
• WP Posts Limit: A critical resource monitor that displays your current usage relative to your plan’s maximum capacity (e.g., 45/100).

This oversight is crucial for users who need to balance their AI’s knowledge depth with technical performance and API costs. Before proceeding with data ingestion, it is recommended to ensure your is correctly set up to handle embeddings.

Diverse Data Source Integration

Antimanual provides a multi-faceted approach to building your Knowledge Base, supporting five primary ingestion methods. This flexibility ensures that regardless of where your documentation or brand information resides, it can be utilized for AI training.

Native WordPress Integration

The most direct method involves selecting existing content from your WordPress installation. This includes standard posts and pages, but also extends to custom post types such as those created by EazyDocs. For more details on this process, see the guide on .

Professional Document Ingestion (Pro)

For organizations with extensive offline documentation, the Pro version allows for direct PDF uploads. The system utilizes advanced text extraction and OCR (Optical Character Recognition) to convert static documents into searchable, interactive AI data. This is ideal for technical manuals, company handbooks, and white papers.

Web Crawling and Snippets

Users can also expand the KB by entering external URLs or manual text snippets. The web crawling engine automatically strips away “noise” like menus and footers, focusing solely on the core content. Manual text snippets are particularly useful for establishing specific rules or short-term guidelines that the AI should follow during conversations.

Resource and Limit Management

Effective management of the Knowledge Base requires an understanding of how items are counted against your limits. Each time a post or document is “Added to KB,” the system generates embeddings via OpenAI. These embeddings are stored locally or in your connected cloud account, allowing for instant retrieval during chatbot interactions or content generation tasks.

Regularly auditing your statistics allows you to remove outdated items or prioritize high-value content, ensuring the AI remains efficient and relevant to your current business objectives.

Frequently Asked Questions