Overview

BrainyFlow does not provide built-in utilities. Instead, we offer examples that you can implement yourself. This approach gives you more flexibility and control over your project's dependencies and functionality.

Available Utility Function Examples

1. LLM Wrapper: Interact with Language Models

2. Visualization and Debugging: Tools for visualizing and debugging flows

3. Web Search: Perform web searches

4. Chunking: Split large texts into manageable chunks

5. Embedding: Generate vector embeddings for text

6. Vector Databases: Store and query vector embeddings

7. Text-to-Speech: Convert text to speech

Why Not Built-in?

We believe it's a bad practice to include vendor-specific APIs in a general framework for several reasons:

1. API Volatility: Frequent changes in external APIs lead to heavy maintenance for hardcoded APIs.

2. Flexibility: You may want to switch vendors, use fine-tuned models, or run them locally.

3. Optimizations: Prompt caching, batching, and streaming are easier to implement without vendor lock-in.

Implementing Utility Functions

When implementing utility functions for your BrainyFlow project:

1. Create a separate file for each utility function in the utils/ directory.

2. Include a simple test or example usage in each file.

3. Document the input/output and purpose of each utility function.

Example structure:

Python

my_project/
├── utils/
│   ├── __init__.py
│   ├── call_llm.py
│   ├── search_web.py
│   └── embed_text.py
└── ...

TypeScript

my_project/
├── utils/
│   ├── callLlm.ts
│   ├── searchWeb.ts
│   └── embedText.ts
└── ...

By following this approach, you can easily maintain and update your utility functions as needed, without being constrained by the framework's built-in utilities.