audio-analyzer

With OpenAI's new skill system for ChatGPT, AI agents can now dynamically load specialized capabilities similar to Claude's skills architecture. This repository adapts the proven Claude skills format to work with ChatGPT, enabling:

• Progressive disclosure: Skills load only when relevant to the current task
• Specialized expertise: Domain-specific workflows and tool orchestration
• Code execution: Python and JavaScript scripts for deterministic operations
• Structured knowledge: Organized instructions, templates, and reference materials

Each skill is a self-contained folder containing:

• SKILL.md: Core instructions with YAML frontmatter (name, description)
• scripts/: Executable Python/Node.js tools for specialized operations
• references/: Supporting documentation, specifications, and examples
• assets/: Templates, configurations, and static resources

When you interact with ChatGPT, the AI evaluates available skills and loads relevant ones dynamically based on your request, similar to how [Claude Skills](https://support.claude.com/en/articles/12512176-what-are-skills) operate.

🪙 Crypto TA Analyzer

crypto-ta-analyzer/

Comprehensive cryptocurrency and stock technical analysis using 29 proven indicators including Bollinger Bands, Ichimoku Cloud, and On-Balance Volume with divergence detection. Features a 7-tier signal system (STRONG_BUY → STRONG_SELL) with volume confirmation, squeeze detection, and divergence warnings.

Key Features (v2.0):

• 29 Indicators: RSI, MACD, BB, OBV, Ichimoku, EMA, SMA, MFI, KDJ, SAR, VWAP, ATR + 17 more
• 7-Tier Signals: STRONG_BUY, BUY, WEAK_BUY, NEUTRAL, WEAK_SELL, SELL, STRONG_SELL
• Divergence Detection: RSI, MACD, and OBV divergence alerts
• Bollinger Squeeze: Low volatility breakout detection
• Volume Confirmation: OBV/MFI agreement scoring
• Generic Data Converter: Auto-detects CoinGecko, Binance, Yahoo Finance, or any OHLCV format
• Regime-Aware: Adapts scoring for trending vs ranging markets

Scripts:

• ta_analyzer.py - Main analysis engine (29 indicators)
• data_converter.py - Generic multi-source data converter with auto-detection

Use for: Price analysis, trading signals, trend identification, divergence detection, breakout setups, multi-indicator technical assessments

📄 Document Converter Suite

document-converter-suite/

Convert between 8 document formats with best-effort text-first extraction. Supports 64 conversion paths (full 8×8 matrix) with single file or batch processing.

Supported Formats:

• Office: PDF, Word (DOCX), PowerPoint (PPTX), Excel (XLSX)
• Text: Plain Text (TXT), CSV, Markdown (MD), HTML

Key Features:

• 64 Conversion Paths: Any format to any format (8×8 matrix)
• Smart Heading Detection: Font size + bold + ALL CAPS heuristics
• Multi-Table PPTX: Creates one slide per table (no dropped tables)
• Data Truncation Warnings: Alerts when XLSX data exceeds limits
• Batch Processing: Recursive directory conversion with pattern matching
• High Fidelity: Markdown ↔ HTML and Markdown ↔ DOCX preserve structure well

Scripts:

• convert.py - Single-file CLI converter
• batch_convert.py - Batch converter for directories

Use for: Document format conversion, content extraction, batch processing, Markdown/HTML to Office, table extraction to CSV/XLSX

🔧 MCP Builder

mcp-builder/

Comprehensive guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with externa