3. Data, Science & AI

Plans synthetic data generation strategies for AI training/testing, addressing data scarcity and restrictions with quality requirements and validation approaches.

Provides structured incident response for AI system failures, including classification, immediate actions, rollback procedures, and post-mortem templates.

Builds proof-of-concept code and demo flows for validated AI capabilities to enable stakeholder experience before production commitment.

Simulates 2D rigid body physics with Box2D API compatibility for LÖVE game development, handling collisions and dynamics.

Manages CLAUDE.md and AGENTS.md files to configure, audit, and synchronize AI coding agent behavior instructions.

Designs evaluation frameworks for LLM applications, including benchmark selection, test case design, and quality thresholds for pre- or post-deployment use.

Shows current status and configuration of an AI inference session.

Transforms vague user intent into actionable roadmaps using coordinated Research, Survey, and Plan agents via multi-agent AI workflow.

Provides system-independent mathematical functions for LÖVE game development, including random number generation and geometric calculations.

Displays help documentation and user guide for the Inference Confidenz AI tool, enabling quick reference to model inference features.

Adjusts the confidence threshold for auto-continuation in AI model responses.

Utility for data transformation, encoding, decoding, and compression operations within LÖVE game development projects.

Organizes and shares AI prompts across teams with version control, quality ratings, and usage guidelines for efficient LLM integration.

Configures AI inference confidence thresholds and model output parameters for optimal prediction reliability.

Provides physics operations, collision detection, and rigid body dynamics for LÖVE game development.

Manages Model Context Protocol (MCP) servers for AI workflows, enabling discovery, execution, and context-aware tool usage across multiple AI capabilities.

Provides CLI workflows for managing Model Context Protocol (MCP) servers, tool interactions, endpoints, and authentication in AI agent development.

Configures settings for the inference-planz plugin, enabling AI model inference configuration.

Provides data processing utilities for encoding, decoding, compression, and transformation in LÖVE game development projects.

Shows current status and recent history of Inference Confidenz for real-time AI model inference monitoring.

Assesses data quality, identifies gaps, and provides remediation recommendations for AI project readiness before commitment.

Manages AI/ML model registries, catalogs metadata, and generates governance reports for model lifecycle tracking.

Assesses whether proposed AI capabilities are achievable with current LLM technology to prevent resource waste.

Provides Python-based web scraping and data extraction for building AI/RAG pipelines, supporting dynamic pages and LLM strategies.