发现 Skills

Evaluate GitHub contributors for MLOps/engineering roles. Use when analyzing candidates, researching GitHub profiles, or updating CONTRIBUTORS.md with hiring assessments.

Riverpod (v3.3.x) compile-safe state management for Flutter with context-free reactive programming and code generation. Use this skill when implementing Riverpod providers with @riverpod annotations, handling AsyncValue for loading/error/data states, using ref.watch/ref.read/ref.listen for provider access, setting up ProviderScope as app root, implementing StateNotifier/StateNotifierProvider, using FutureProvider/StreamProvider for async data, enabling auto-retry and auto-dispose mechanisms, utilizing family/autoDispose modifiers, implementing provider dependencies and combining providers, migrating from Provider to Riverpod, debugging ProviderNotFoundException errors, testing providers in isolation, preventing memory leaks with autoDispose and ref.onDispose, avoiding UnmountedRefException in async operations, implementing lifecycle-aware state updates with ref.mounted checks, or choosing between ref.read and ref.watch for optimal rebuild performance. Ideal for apps requiring compile-time safety, testable state without BuildContext dependency, global state access from anywhere, complex async data flows, memory leak prevention, or avoiding ChangeNotifier performance pitfalls.

Self-contained app generator — invoke this skill directly, do not decompose into sub-steps. Generates React web apps with the Fireproof database. Use when creating new web applications, adding components, or working with local-first databases. Ideal for quick prototypes and single-page apps that need real-time data sync.

Comprehensive R package for command-line interface styling, semantic messaging, and user communication. Use this skill when working with R code that needs to: (1) Format console output with inline markup and colors, (2) Display errors, warnings, or messages with cli_abort/cli_warn/cli_inform, (3) Show progress indicators for long-running operations, (4) Create semantic CLI elements (headers, lists, alerts, code blocks), (5) Apply themes and customize output styling, (6) Handle pluralization in user-facing text, (7) Work with ANSI strings, hyperlinks, or custom containers. Also use when migrating from base R message/warning/stop, debugging cli code, or improving existing cli usage.

Ensures all commit messages and PR titles follow the Conventional Commits specification.

Conduct thorough academic peer reviews with structured feedback using load_pdf and arxiv_to_prompt. Use when the user asks to review a paper, provide manuscript feedback, critique research, or write a referee report.

Routes a complex or ambiguous task through the right subagent and ensures all required skills/agents are installed AND up-to-date before dispatching. Reads ~/.claude/CLAUDE-agents.md and ~/.claude/CLAUDE-skills.md, audits local components, checks upstream versions, and proposes installs/updates with a single consolidated approval prompt. Use when the task spans 2+ domains, when you do not know which skill/subagent to pick, or when you want to guarantee fresh components before execution. v0.2.0 adds a 3-mode recommendation (INSTALL_BEST / INSTALL_MULTIPLE / SYNTHESIZE) via analyze-candidates and the FTS5 deep index.

QCSD Development phase swarm for in-sprint code quality assurance using TDD adherence, code complexity analysis, coverage gap detection, and defect prediction. Consumes Refinement outputs (BDD scenarios, SFDIPOT priorities) and produces signals for Verification.

Review system costs and propose reductions. Use when a senior developer needs cost-saving recommendations.

Крипта: курсы BTC, анализ, новости, мониторинг цены

Twitch: статус стримеров, кто онлайн

CI-only self-improvement workflow using gh-aw (GitHub Agentic Workflows). Captures recurring failure patterns and quality signals from pull request checks, emits structured learning candidates, and proposes durable prevention rules without interactive prompts. Use when: you want automated learning capture in CI/headless pipelines.

Fast-forward through OpenSpec artifact creation. Use when the user wants to quickly create all artifacts needed for implementation without stepping through each one individually.

YASB (Yet Another Status Bar) - A highly customizable Windows status bar with widgets, CSS styling, keybindings, and Python plugin development. Use when configuring YASB status bar, creating custom widgets, styling with CSS, setting up keybindings, managing via CLI, or troubleshooting YASB issues. Keywords: yasb, status-bar, windows, configuration, yaml, css-customization, widgets, keybindings, hotkeys, dark-mode, nerd-fonts, cli, python, pydantic, qt-framework, autostart, troubleshooting, environment-variables, widget-development, desktop-theming.

Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", "review UX", or "check my site against best practices".

Analyze log files by stripping ANSI escape sequences first. Use when asked to process, handle, read, or analyze log files that may contain terminal escape codes.

This skill should be used when implementing security for Stripe webhook endpoints, handling "webhook rate limiting", "Stripe secret management", "webhook abuse prevention", "log redaction for billing", "STRIPE_WEBHOOK_SECRET storage", "webhook endpoint protection", "Stripe API key security", "billing security review", "sensitive data in logs", "production webhook hardening", "DDoS protection for webhooks", or when the user asks about securing Stripe webhook endpoints and billing infrastructure in Fastify.

Enforce a “docs-first, code-second” approach for OpenClaw configuration and behavior questions. Use this when answering questions about OpenClaw config options, routing, precedence, CLI behavior, outages, or implementation details. Prefer official documentation; when behavior is unclear or depends on version, verify in the source code using ripgrep (rg). If the local repository or tools are unavailable, fall back to asking the user to run exact copy-paste commands locally.

Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.

[Implementation] Thorough implementation with maximum verification

Create OpenSCAD files for complex 3D geometry in URDF, leveraging available libraries and automatic STL conversion

Generate highly detailed, Midjourney-style image prompts optimized for the FLUX 1.1 Pro model on Replicate. Transform basic user descriptions into rich, cinematic prompts with professional photography qualities, dramatic lighting, and editorial-quality aesthetics. Use when users request image generation, need prompt enhancement, or want Midjourney-quality outputs via FLUX 1.1 Pro.

Operate Burp Suite Community and Professional for web application penetration testing. Use when intercepting HTTP/S traffic, fuzzing parameters, testing authentication bypasses, scanning for vulnerabilities, or performing advanced web application assessments. Covers proxy setup, Intercept, Repeater, Intruder (all attack types), Scanner, Sequencer, Decoder, Comparer, Logger, Collaborator, BApp Store extensions (Autorize, Param Miner, JWT Editor, Turbo Intruder, ActiveScan++), scope control, session handling, macros, REST API, and Bambda filters.

Help users synthesize and act on customer feedback. Use when someone is analyzing NPS responses, processing support tickets, reviewing user research, synthesizing feedback from multiple channels, or trying to identify patterns in customer input.