Browse reusable Agent Skills, each with a clear purpose and practical guidance.
500+ item EVM smart contract audit system across 19 domains. Runs parallel specialist agents, synthesizes findings, and files GitHub issues. Separate from the security skill — this is for auditing contracts you did not write.
Challenge causal claims through structured threat assessment, counterfactual reasoning, and CausalPy falsification checks. Use when validating whether a causal effect is real or when the user asks "is this effect real?" or "can I trust this result?"
DART Verify Sim: text-first and visual checks for 3D scenes and physics (metrics, scene dump, trajectories, headless render, image verdict/golden)
Help an engineer author, review, or pre-merge-gate a pull request to a ROCm library so it is traceable, tested, and safe to merge. Use when preparing a PR, reviewing a PR, or deciding whether an approved PR is safe to merge right now, or when the user provides a GitHub PR URL or branch and asks for help with PR quality, description, testing, review, or merge-readiness. Library-agnostic base; component overlays extend it.
Picks and runs the examples under examples/ that exercise the code changed on the current branch / PR / diff, as a phase-two integration check after a code review. Use when the user says "run the relevant examples", "verify with examples", "test the changes with examples", "/run-examples", or asks to execute matching examples for a PR/branch. Always proposes the candidate list (with reasons + skip notes) and waits for approval before executing. Handles missing API keys via examples/.env.local and brings up the right Docker services when needed.
TheRock build-repo supplements to the ROCm PR quality base skill. Use for TheRock PR author, review, or pre-merge gating where the change touches the superbuild, submodules/patches, artifact descriptors, or reusable CI workflows. Adds and tightens base rules; never relaxes a base MUST.
Use when you need to visually verify macOS native app UI after fixing bugs, without asking the user to manually test. Triggers when: a UI fix is implemented and needs screenshot-based self-validation; verifying that a visual bug is actually gone (e.g. focus ring, wrong color, missing indicator); closing a feedback loop on Settings/window UI changes in FluxMarkdown or similar macOS apps. Use proactively after ANY SwiftUI or AppKit UI change.
Analyze Drift / SQLite slow-query and super-slow-query logs against this app's known stall patterns (read waves, N+1, transaction scoping, MultiExecutor contention, WAL/OS factors)
Act as a skeptical senior engineer performing a detailed code review of the latest changes on the current branch (or a given PR) — best practices, maintainability, performance, security, and readability. Objective and concise; only real issues, with a clear "no issues" verdict when the code is clean. Replaces CodeRabbit / Gemini Code Review.
Diagnose common DGX Station GB300 issues — CUDA crashes, wrong-GPU targeting, vLLM/SGLang container bugs, MIG state problems, NVLink/Fabric Manager errors, X/Vulkan failures, HuggingFace auth, and port conflicts. Use when the user reports a GPU error, inference server crash, MIG problem, or any unexplained DGX Station failure.