Agent skills for ChatGPT

Load this skill and follow it when optimizing compiled graphs, fusion, graph breaks, recompilation, GPU graphs, or multi-GPU runtime…

Use when processing large PDFs, document collections, or bulk text extraction tasks that benefit from GPU-accelerated processing.

Use when working with GPU programming, Metal shaders, Apple Silicon GPU architecture, or any GPU/graphics topic.

Load this skill and follow it when placing or moving GPU resources across device memory, peer devices, host memory, storage, or remote…

Load this skill and follow it when optimizing GPU code that may change floating-point evaluation order, precision, boundary behavior, NaN…

Load this skill and follow it when turning GPU optimizations into reviewable patches, designing guards and fallbacks, running…

Load this skill and follow it when establishing a GPU performance baseline, analyzing profiler data, roofline results, or hardware…

Load this skill and follow it when designing runtime GPU state that survives across kernels, steps, requests, or sessions, especially when…

Load this skill and follow it when optimizing GPU reductions, scans, prefix operations, recurrences, streaming state, tile-local partials…

Build, debug, or review WebGPU renderer work — three.js/TSL scene layers, node materials, or raw WGSL passes and compute.

Load this skill and follow it when optimizing GPU resource lifetimes, allocation reuse, transient aliasing, pooling, workspace capacity…

Load this skill and follow it when diagnosing GPU memory allocatability failures, internal or external fragmentation, virtual-to-physical…

Answer questions about a GRACE 4 project by navigating .grace current-state artifacts and file-local semantic markup.

Debug and fix issues in a GRACE 4 project using .grace semantic navigation, assertions, and verification evidence.

Interview the user and create an approved GRACE 4 GraceChangeSpec plus optional design-context.xml inside .grace/changes/active/C-*/.

Trigger GRADE quality assessment automatically when new research sources or findings enter the corpus

Score any draft post against the grading rubric (hook 50%, voice 15%, length 10%, CTA 10%, platform-native 10%, anti-AI 5%).

Generate corpus-wide GRADE quality distribution report

Build Gradio web UIs and demos in Python.

Debug and optimize Android/Gradle build performance.

Collects, verifies, ranks, and tracks campus recruitment and graduate job opportunities with structured status monitoring.