3. Data, Science & AI

Enables deep learning model training and iterative processing through the deep-loop plugin.

Enables MySQL database operations (CRUD) via Python and Node.js scripts for data management and manipulation.

Estimates YouTube Shorts revenue using Python/Node.js scripts based on view counts and regional RPM metrics.

Prevents AI from generating incorrect data by halting and requesting clarification when uncertain, ensuring response accuracy.

Generates images via Gemini models with text-to-image, image-to-image, and multi-image composition capabilities.

Generates images using Antigravity API as primary, automatically falling back to NanoBanana on failure. Supports custom dimensions, quality levels, and batch processing.

Analyzes AI-generated images/videos to reverse-engineer input prompts, aiding in prompt optimization and style learning.

Enables non-interactive Kimi LLM integration in Claude Code with session management, context injection, and agent role support.

Analyzes user data in info/ directory to generate structured profiles, supporting creation, refresh, and automatic updates.

Provides AI-powered image generation via AllAPI's NanoBanana models, supporting text-to-image, image-to-image, and multi-image synthesis with Pro and Flash variants.

Expert guidance for Supabase database design, SQL query optimization, authentication implementation, and real-time feature setup in production applications.

Expert in crafting effective prompts, fine-tuning LLMs, and building RAG systems for scalable AI applications.

Non-interactive Gemini AI collaboration in Claude Code with session management, context injection, and role-based agent workflows.

Expert assistance for database architecture, schema design, query optimization, and caching strategies across PostgreSQL, MySQL, MongoDB, and Redis at production scale.

Batch-generates images via OpenAI API with random prompts and outputs an HTML gallery for display.

Structures data files in supported formats to enhance Claude's reasoning and extraction capabilities, resolving context window constraints.

Enables offline text-to-speech synthesis using the sherpa-onnx framework, processing audio locally without cloud dependency.

AI agent orchestrator that delegates tasks to specialized agents in parallel, validates outputs, and ensures high-quality results through coordinated execution.

Provides a command-line interface for Gemini AI to perform one-shot Q&A, text summarization, and content generation tasks.

Local speech-to-text conversion using OpenAI's Whisper model via CLI, requiring no API key or internet connection.

Guides efficient SQL Server database schema design and query optimization for improved data management and performance.

Generates spectrograms and feature-panel visualizations from audio files using the songsee command-line tool for data analysis and audio feature exploration.

Captures frames or clips from RTSP/ONVIF cameras for data acquisition in video analytics and monitoring systems.

Transcribes audio files to text using OpenAI's Whisper API for speech recognition tasks.