Loading...
Loading...
Found 2,163 Skills
Defense-in-depth verification before declaring any task complete. Run tests, check build, validate changed files, verify no regressions. Applies 4-level adversarial artifact verification (EXISTS > SUBSTANTIVE > WIRED > DATA FLOWS) with goal-backward framing. Use before saying "done", "fixed", or "complete" on any code change. Use for "verify", "make sure it works", "check before committing", or "validate changes". Do NOT use for debugging (use systematic-debugging) or code review (use systematic-code-review).
Expert knowledge for Azure Analysis Services development including troubleshooting. Use when testing server connections, debugging gateway or firewall blocks, or checking connection strings and ports, and other Azure Analysis Services related development tasks. Not for Azure Synapse Analytics (use azure-synapse-analytics), Azure SQL Database (use azure-sql-database), Azure SQL Managed Instance (use azure-sql-managed-instance), SQL Server on Azure Virtual Machines (use azure-sql-virtual-machines).
Correctly call native (C/C++) libraries from .NET using P/Invoke and LibraryImport. Covers function signatures, string marshalling, memory lifetime, SafeHandle, and cross-platform patterns. USE FOR: writing new P/Invoke or LibraryImport declarations, reviewing or debugging existing native interop code, wrapping a C or C++ library for use in .NET, diagnosing crashes, memory leaks, or corruption at the managed/native boundary. DO NOT USE FOR: COM interop, C++/CLI mixed-mode assemblies, or pure managed code with no native dependencies.
Enforces spec-before-code workflow for AI-driven development. Automatically selects Spec-Kit or OpenSpec mode, triages complexity (quick/standard/thorough), recovers session context, and applies quality gates (G0-G4) with automated review loops at every stage. Use this skill whenever the user says "/super-spec", "spec first", "规范先行", or starts any feature, bugfix, or refactor — especially in projects with .spec-mode, .specify/, or openspec/ directories. Even if the user doesn't explicitly ask for spec-driven workflow, activate this skill for any non-trivial code change to prevent skipping the design phase. Orchestrates: Spec-Kit/OpenSpec (OPSX) + planning-with-files + ui-ux-pro-max (v2.0, 67 styles, 161 palettes, 13 stacks) + Superpowers (TDD, code review, verification, debugging, spec/plan review loops, subagent model selection).
Uncle Huang's Private Advisory Board — a business decision think tank composed of 12 top thinkers. With a structured private advisory board process, different cognitive frameworks collide to produce optimal decisions. Trigger scenarios: - Users say "start a private advisory board", "invite the think tank", "help me make decisions" - Users say "Private Advisory Board: [topic]", "Let the experts discuss this matter" - Users use the /私董会 or /advisory-board command - Users face major business decisions requiring multi-perspective analysis Non-trigger scenarios: - Simple operational issues (how to publish, how to typeset) - Pure technical issues (code debugging) - Daily content creation (use article-writer / topic-partner) - Personal emotional counseling (use life-coach)
Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics. Covers training loops, optimizer selection (AdamW, Muon), LR scheduling, mixed precision, debugging, and systematic experimentation. Use when training or fine-tuning neural networks, debugging loss spikes or OOM, choosing architectures, or optimizing GPU throughput.
Manages iOS Simulator devices and tests app behavior using xcrun simctl. Covers device lifecycle (create, boot, shutdown, erase, delete), app install and launch, push notification simulation, location simulation, permission grants via privacy subcommand, deep link testing via openurl, status bar overrides, screenshot and video recording, log streaming with os_log filtering, get_app_container paths, and #if targetEnvironment(simulator) compile-time checks. Use when creating or managing simulator devices, testing push notifications without APNs, simulating GPS locations, granting or resetting privacy permissions, capturing screenshots or screen recordings from the command line, streaming device logs, debugging simulator boot failures, troubleshooting CoreSimulator issues, or checking simulator hardware limitations.
Browser automation and testing using chrome-devtools MCP server. Use when automating web browsers, taking screenshots, inspecting console logs, monitoring network requests, testing responsive layouts, collecting performance metrics, or debugging web applications. Critical for visual testing workflows and browser-based automation tasks.
Master end-to-end testing to build reliable test suites that catch bugs, improve confidence, and enable fast deployment. Use when implementing E2E tests, debugging flaky tests, or establishing testing standards.
Capture the current task into a structured temporary session bundle under `.agents/sessions/` so a learning agent can later distill durable repo knowledge. Use for completed, blocked, or abandoned tasks with meaningful changes, debugging, validation, or reusable lessons.
Use to help users get started with Nemo Gym reward profiling. Covers the basic ng_run, ng_collect_rollouts, and ng_reward_profile workflow, repeated rollouts, materialized inputs, rollout JSONL artifacts, task and rollout identity, output inspection, partial profiling, and rollout_infos. For failed jobs, prefer nemo-gym-debugging.
Use when adding, modifying, optimizing, or debugging CuTile autotuning code. Trigger signals: `exhaustive_search` / `replace_hints` / `hints_fn` / `cuda.tile.tune` in code, `autotune` in filenames, or correctness/performance issues in autotuned CuTile kernels. Covers: tune-once/cache/launch pattern, per-architecture configs (sm80–sm120), parameter space design (tile sizes, occupancy, num_ctas), and 7 common pitfalls with solutions.