3. Data, Science & AI

Detects and assesses AI model capabilities to ensure optimal cross-model compatibility and deployment efficiency.

Provides advanced contextual pattern recognition, semantic similarity analysis, and cross-domain pattern matching for enhanced AI learning capabilities.

Enables autonomous pattern recognition, storage, and retrieval with self-learning for continuous project-level improvement.

Anticipates and pre-loads optimal skills before task execution using historical data analysis and pattern recognition to enhance system performance.

Analyzes conversation mistakes by correlating them with conversation length to identify potential causes.

Provides a minimal CLI interface for executing additional AI models directly in a terminal with TTY support.

Guide for efficiently running Qwen3-TTS (VoiceDesign mode) on Mac using MLX, featuring parameter optimization, RTF tuning, and Remotion-integrated Web UI construction.

Guide for efficient Qwen3-TTS (VoiceDesign mode) execution on Mac MLX, including parameter optimization, RTF tuning, and Remotion Web UI integration.

Orchestrates multi-model AI deliberation (Claude, Codex, Gemini) with availability checks and structured output generation.

Launches a subagent within an isolated Git worktree, preserving task context and tracking token usage for efficient LLM-based agent workflows.

Generates comprehensive token usage reports with threshold analysis and actionable recommendations for optimizing AI model consumption.

Collects results from completed AI subagents, including code commits, metrics, and state updates for system monitoring.

Monitors AI subagent status, including token consumption and context window limits for optimized LLM resource management.

Coordinates parallel execution of multiple AI subagents with synchronized result aggregation for efficient LLM agent workflows.

Compares two documents semantically while preserving structural relationships to highlight content and structural differences.

Creates and maintains example implementations for AI SDK functions to validate provider compatibility, demonstrate features, and serve as test fixtures.

Processes Excel files and generates reports for data analysis. Use for spreadsheet data transformation and reporting.

Generates optimized handoff prompts for delegating tasks between LLM agents, ensuring clear context transfer for sub-tasks or new agent contexts.

Automates web page asset extraction, theme/data/document recovery, and page restoration using Axhub and Firecrawl MCP workflows.

Generates Claude Code subagents with structured YAML frontmatter, system prompts, and tool configurations for AI agent orchestration.

Creates new AI agent skills following best practices and documentation, enabling users to define custom capabilities via natural language prompts.

Generates comprehensive LLM application blueprints with prompt design, evaluation plans, safety architecture, and launch checklists for RAG and agent systems.

Creates comprehensive LLM evaluation packs with test sets, rubrics, and iteration loops for quality assessment and deployment decisions.

Assesses and mitigates bias in machine learning models using Fairlearn, providing compliance reports for fairness.