258 skills · sorted by relevance
Support manufacturing quality investigations from containment through root cause analysis and corrective action. Organize decisions about nonconforming materials, assess whether actions prevent recurrence, interpret process control charts, and address supplier quality problems.
Review decorative images and icons to prevent unnecessary screen reader announcements. Check empty alt text, aria-hidden, and relevant keyboard behavior, then suggest fixes for concrete accessibility issues without flagging decorative assets solely because they exist.
Measure CLS, a score for unexpected movement of page content, and identify what causes it. Review image dimensions, reserved advertising space, fonts, and animations to recommend changes that reduce reading interruptions and accidental clicks.
Review code that manages the conversation history and extra information sent to Codex models. Apply rules for incremental history, size limits, cache stability, manual review of large additions, and required implementation structures.
Improve Remotion Studio rendering performance by limiting which components update with every frame. The skill guides on-demand timeline reads and React.memo usage, helping keep stable components from rerendering unnecessarily while preserving updates where the displayed frame matters.
Validate and convert crystal and molecular structures, compare symmetry assignments, and build phase diagrams showing relative stability from local calculation data. Preserve sources and parameters while analyzing materials data or planning limited Materials Project queries.
Prepare and run molecular dynamics simulations with OpenMM, then analyze molecular motion with MDAnalysis. Examine protein flexibility, structural changes, and protein–ligand contacts through calculations and plots of simulated trajectories.
Simulate a uniform gas mixture in a closed, insulated reactor at constant volume or pressure. Calculate ignition delay from the peak temperature rise rate, export temperature and species histories, and check conservation and numerical stability.
Review AI skills and bundles without running their code. Checks for data leaks, suspicious permissions, hidden code, and system changes across desktop and mobile platforms, then reports flagged actions, risk scores, and suggested mitigations.
Prepare GRPO reinforcement learning configurations for tasks with programmatically checkable answers, such as math, code, and structured output. Design and inspect reward functions, then select training variants when runs show instability or exploit flaws in scoring.
Compare job offers using salary, equity, benefits, and working conditions. Build side-by-side tables for first-year and ongoing compensation, then score career growth, team culture, and other factors according to personal priorities.
Plan and configure large language model training with OpenRLHF to better match human preferences. Covers training method selection, distributed GPU setup, inference acceleration with vLLM, and troubleshooting memory or training stability issues.
Guides large language model reinforcement learning with miles, covering environment setup, low-precision training, consistency between training and inference, and speculative decoding, where a smaller model proposes tokens for a larger model to verify.
Build and train Mixture of Experts models, which activate selected specialist networks for each input. Use DeepSpeed or HuggingFace workflows to configure routing, balance workloads, distribute experts across GPUs, and optimize inference.
Diagnose failing or intermittent Playwright tests by reproducing errors and examining execution traces. Classify the cause, apply a targeted fix, and run the test ten times to check stability, with recommendations to prevent recurrence.
Guide substantial Swift changes in cmux with rules for package boundaries, dependency direction, state management, concurrency, and API documentation. Structure code so public package types can be tested without launching the app.
Prepare a long play session with timed observation checkpoints and logs for memory growth, stability, and player fatigue. It creates the testing protocol and analysis template; a human plays the game and records the findings.
Track tests that pass and fail on unchanged code using GitHub CI history. Score their instability over multiple runs, propose reasoned skips in a pull request, maintain a tracking issue, and summarize findings in Slack.
Investigate service errors in Sentry by project, environment, and time period. Rank problems by event counts and affected users, compare releases or earlier periods, and report findings with links and a clear search scope.
Sets up AFL++ to test C/C++ programs with varied generated inputs, a technique called fuzzing. Covers running tests across CPU cores, monitoring progress, reducing input collections, and reproducing crashes for investigation.
Improve a web page’s loading speed, response to user actions, and visual stability. The skill guides targeted fixes for LCP, INP, and CLS, with code examples, measurement methods, and checklists for each metric.
Add, repair, and strengthen tests for existing code using the project's testing framework. Covers regression cases, boundary conditions, flaky tests, and coverage gaps, with validation that links requested behaviors to specific tests.
Make specific C# behavior testable by adding a minimal way to replace time, randomness, or other external dependencies. Preserve public APIs and production behavior, then run repeatable tests without real file access or waiting.
Configure dependency injection in .NET MAUI, a way to supply objects with the services they need. Covers service registration, instance lifetimes, page and ViewModel constructors, Shell navigation, platform implementations, and test substitutions.