Reviews, validates, and safely merges Dependabot pull requests for the sq.io site (site/, Bun lockfile). Use when clearing site dependency PRs, triaging Dependabot failures, or checking Lighthouse impact before merge.

Reviews and merges Dependabot pull requests for Go modules (gomod) at the sq repo root. Use for dependabot gomod PRs, go.mod/go.sum updates, and Go module security bumps—not site/ Bun PRs.

Reviews and merges Dependabot pull requests for GitHub Actions (the github-actions ecosystem) that bump `uses:` pins in `.github/workflows/`. Use for Dependabot github_actions PRs (branches like `dependabot/github_actions/...`), not go.mod or site/ Bun PRs.

Generate, edit, and read PowerPoint presentations. Create from scratch with PptxGenJS (cover, TOC, content, section divider, summary slides), edit existing PPTX via XML workflows, or extract text with markitdown. Triggers: PPT, PPTX, PowerPoint, presentation, slide, deck, slides.

Address review comments (including Copilot comments) on the active pull request. Use when: responding to PR feedback, fixing review comments, resolving PR threads, implementing requested changes from reviewers, addressing code review, fixing PR issues.

Use this skill when deploying standalone RT-VLM dense captioning or calling its REST API (uploads, captions, streams, chat-completions, Kafka). Not for VSS profile deploy or video-search ingestion.

Iteratively optimize cuTile kernel performance through systematic profiling, bottleneck analysis, IR comparison, and targeted tuning. Covers tile sizes, occupancy, autotune configs, TMA, latency hints, persistent scheduling, num_ctas, flush_to_zero, and IR-level debugging. Use when asked to "optimize cutile kernel", "improve kernel perf", "tune cutile performance", "make kernel faster", or iteratively benchmark and refine a cuTile GPU kernel in the TileGym project.

Monocular depth estimation using Metric Depth Anything v2 or Relative Depth Anything architectures. Predicts per-pixel depth from single RGB images. Use when training, evaluating, exporting, or running inference for a TAO monocular depth model. Trigger phrases include "train monocular depth", "DepthAnything v2", "metric depth from single image", "monocular depth estimation".

Multi-step video annotation pipeline that turns raw videos into Chain-of-Thought training data — multi-level captions, structured descriptions, and QA pairs (MCQ, binary, open-ended) with reasoning traces, via VLM/LLM distillation. Use when the user wants to "create video training data", "generate video QA datasets", "build CoT reasoning traces from videos", "auto-label videos", or run the video_reasoning_annotation pipeline. Triggers include "video annotation", "video CoT", "video QA", "chain-of-thought", "video captioning pipeline", "video distillation".

Validate and use CPU offloading in Megatron Bridge, including layer-level activation offloading and fractional optimizer state offloading with HybridDeviceOptimizer.

Use when asked about Studio job analytics — compute hours, user spend, failure rates, cost estimation, cluster usage. Generates and maintains dc-knowledge/jobs/index.md.

This skill helps an LLM pick the right AxAgent context tool for a job - contextMap for recurring corpora, contextPolicy presets for within-run trajectory compaction, agent.optimize for offline GEPA instruction/demo tuning, agent.playbook for an evolving context playbook (offline evolve + online update), and recall/memories + skills for per-turn retrieval. Use when the user asks "which context feature should I use", confuses contextMap with contextPolicy or memory, or wants a decision guide for long-context agents. For contextPolicy/contextMap codegen use ax-agent-rlm; for recall/skills use ax-agent-memory-skills; for agent.optimize or agent.playbook use ax-agent-optimize.

Use when writing Python code with `axllm` for flows, nodes, program graphs, nested programs, dynamic options, caching, and optimizer components.

Diagnoses and fixes .NET MAUI development environment issues. Validates .NET SDK, workloads, Java JDK, Android SDK, Xcode, and Windows SDK. All version requirements discovered dynamically from NuGet WorkloadDependencies.json — never hardcoded. Use when: setting up MAUI development, build errors mentioning SDK/workload/JDK/Android, "Android SDK not found", "Java version" errors, "Xcode not found", environment verification after updates, or any MAUI toolchain issues. Do not use for: non-MAUI .NET projects, Xamarin.Forms apps, runtime app crashes unrelated to environment setup, or app store publishing issues. Works on macOS, Windows, and Linux.

Use when writing Rust code with `axllm` for named deployment profiles, generic provider clients, model selection, OpenAI-compatible calls, Responses, Gemini, Anthropic, routers, and balancers.

Use when writing C++ code with `axllm` for AxGen programs, forward calls, indexed multi-sampling, result pickers, streaming, tools, assertions, traces, usage, and output parsing.

Sparse Identification of Nonlinear Dynamics (SINDy) — discover governing equations from time-series data. Builds sparse dynamical system models dx/dt = f(x) from measurements using PySINDy. Use when you have trajectory data and want to find the underlying ODE.

Solve ordinary differential equations (initial and boundary value problems). Supports stiff/non-stiff systems, event detection, Hamiltonian/symplectic integration, parameter sweeps, and phase space analysis. Use for any ODE system in physics, engineering, or applied math.

Track and visualize ML training experiments with Trackio. Use when logging metrics during training (Python API) or retrieving/analyzing logged metrics (CLI). Supports real-time dashboard visualization, HF Space syncing, and JSON output for automation.

Rigorous methodology for evaluating ML models on established benchmarks. Covers proper train/val/test splits, baseline verification from original papers, exact metric formula discrepancies, data-leak detection checklist, multi-seed robustness, and honest reporting templates. Use when claiming to beat published baselines, writing methods papers, or auditing existing results.