Agent
Look for blind spots, unsupported architecture families, missing validation gates, and overclaimed optimizations.
MLX Porting Toolkit — an agent-guided, evidence-gated pipeline (scaffold → convert → parity → benchmark) plus a portable skill for porting PyTorch/Hugging Face models to Apple MLX.
Agent
Look for blind spots, unsupported architecture families, missing validation gates, and overclaimed optimizations.
Agent
Sample model cards, library tags, downloads, tasks, and license metadata to identify porting demand and artifact shapes.
Agent
Find MLX or Apple-supported APIs, runtime constraints, packaging rules, and validation gates that should constrain porting guidance.
Agent
Track packages and release metadata that change install, conversion, runtime, or optional dependency guidance.
Agent
Find architecture-specific algorithm constraints and evaluation methods for model families the skill supports or should support.
Agent
Mine implementation writeups and practitioner reports for leads while keeping unsupported claims out of supported guidance.