Rodrigo Rodriguez (Pragmatismo) 3bb19a3b4c Add comprehensive documentation for GeneralBots, including keyword references, templates, and user guides

- Created detailed markdown files for keywords such as HEAR, TALK, and SET_USER.
- Added examples and usage notes for each keyword to enhance user understanding.
- Developed templates for common tasks like enrollment and authentication.
- Structured documentation into chapters covering various aspects of the GeneralBots platform, including gbapp, gbkb, and gbtheme.
- Introduced a glossary for key terms and concepts related to GeneralBots.
- Implemented a user-friendly table of contents for easy navigation.

2025-10-25 14:50:14 -03:00

1.9 KiB

Raw Blame History

.gbkb Knowledge Base

The .gbkb package manages knowledge base collections that provide contextual information to the bot during conversations.

What is .gbkb?

.gbkb (General Bot Knowledge Base) collections store:

Document collections for semantic search
Vector embeddings for similarity matching
Metadata and indexing information
Access control and organization

Knowledge Base Structure

Each .gbkb collection is organized as:

collection-name.gbkb/
├── documents/
│   ├── doc1.pdf
│   ├── doc2.txt
│   └── doc3.html
├── embeddings/          # Auto-generated
├── metadata.json       # Collection info
└── index.json         # Search indexes

Supported Formats

The knowledge base can process:

Text files: .txt, .md, .html
Documents: .pdf, .docx
Web content: URLs and web pages
Structured data: .csv, .json

Vector Embeddings

Each document is processed into vector embeddings using:

BGE-small-en-v1.5 model (default)
Chunking for large documents
Metadata extraction and indexing
Semantic similarity scoring

Collection Management

Creating Collections

ADD_KB "company-policies"
ADD_WEBSITE "https://company.com/docs"

Using Collections

SET_KB "company-policies"
LLM "What is the vacation policy?"

Multiple Collections

ADD_KB "policies"
ADD_KB "procedures"
ADD_KB "faqs"
REM All active collections contribute to context

Semantic Search

The knowledge base provides:

Similarity search: Find relevant documents
Hybrid search: Combine semantic and keyword
Context injection: Automatically add to LLM prompts
Relevance scoring: Filter by similarity threshold

Integration with Dialogs

Knowledge bases are automatically used when:

SET_KB or ADD_KB is called
Answer mode is set to use documents
LLM queries benefit from contextual information

1.9 KiB Raw Blame History