В каталоге 183 730 навыков в 14 категориях.
Spec-driven task management for features and breaking changes using OpenSpec format. Use when creating new features, pla...
TypeScript development with strict mode, Vitest testing, ESLint linting, and CI/CD best practices. Use when working on T...
Expert guidance for writing UP SHIP! rulebook content following Euro-style board game best practices. Use when drafting...
Unified AI agent configuration propagation across 18+ coding assistants.
Use when the user asks about lfg/GrokBuild rules injection, project rules files, matching, or environment configuration.
Discover and apply Overskill skills for this project
Postgres connections are expensive (1-3MB RAM each). Without pooling, applications exhaust connections under load.
Use when the user asks about Codex Rules behavior, injected project rules, supported rule file locations, matching, or e...
A skill is an area of specialization associated with an ability check. If you have [[proficiency]] in a skill, you can a...
A test skill for E2E conversion testing
AI-First SDLC compliance rules and standards. Use when checking code quality, validating architecture, enforcing zero te...
Explain which rules, AGENTS.md files and skills agy loads, in what order, and how to settle conflicts. Use when guidance...
The mandatory rules R-01 to R-38 in three groups: hard gate (absolute), purpose-gate (technique allowed, reason required...
Add a rule to the set of rules
Use when the user wants to create, edit, list, or manage Claude Code rules in .claude/rules/. Provides guided elicitatio...
Scan skills to extract cross-cutting principles and distill them into rules — append, revise, or create new rule files
Scan skills to extract cross-cutting principles and distill them into rules — append, revise, or create new rule files....
Distill principles from the installed skills into the always-loaded rules layer.
Scan .claude/rules/ for broken paths, coverage gaps, and cross-reference issues
Evaluate Claude Code rules in .claude/rules/. Use for frontmatter, globs, and quality audits.
Use after the tech-stack-confirmation skill has been completed and a confirmed tech stack document exists. This skill re...
Settle a board game rules dispute like a fair judge — reconstruct the situation, rule from the rulebook text (pasted or...
检查和更新 Claude Rules 内容。包括格式检查、统计数据同步、过时内容清理。使用 /rules-cleanup 或 /rules-update 调用。
MIGRATE CLAUDE.md into modular `.claude/rules/` directory structure following Claude Code's rules system. Converts monol...
Optimize and create Claude Rules (.claude/rules/*.md) following best practices.
Organize Claude Code instructions into modular rule files in .claude/rules/. Use when setting up project standards, addi...
Audit ~/.claude/rules for residency cost, staleness, redundancy and substrate absorption, and give each rule a verdict....
Compresses and merges CLAUDE.md and rules files to fit under the 40K character cap
Iterative rubric-based evaluation and self-improvement loop. Builds a scoring rubric interactively, evaluates an artifac...
Embody Rumi - AI persona expert with integrated methodology skills
Ruminant-specific medicine covering cattle, sheep, and goat clinical approaches. Four-compartment forestomach physiology...
Generate rumor artifacts that cross-link items/quests/regions and seed rabbit holes.
Build, install, and run the app on a target device (auto-detects platform)
Run a single experiment iteration. Edit the target file, evaluate, keep or discard.
One-shot lifecycle command that chains init → baseline → spawn → eval → merge in a single invocation.
Sustained metric-improvement loop with atomic commits, auto-rollback, and experiment logging. Iterates with specialist a...
러닝 훈련 데이터를 분석하고 Bakken 노르웨이 모델 기반 코칭 피드백을 제공합니다
Run a full pipeline for a task. Orchestrates roles through stages (standalone or HOTL-integrated).
Run TypeScript solution against test cases. Use when user wants to test their code.
One-shot lifecycle command that chains init → baseline → spawn → eval → merge in a single invocation. Use when the user...
Write a SLICE's run lens as a narrative grounding doc (run.md) AND a machine-readable per-environment definition (run.ya...
Execute a named script defined in .aiwg/aiwg.config or list all available scripts
Run the Keychat app on a device or simulator
The universal execution primitive for AI agents. A secure, sandboxed environment designed to compile, execute, and mana...
Build and launch GridRacer in the iOS Simulator
Run a single experiment iteration. Edit the target file, evaluate, keep or discard. Use when the user runs /ar:run or as...
Use — One-shot lifecycle command that chains init → baseline → spawn → eval → merge in a single invocation.
Identify and run only the test suites affected by the current change set. Triggers on phrases like 'run relevant tests',...
End-to-end Coval accent testing workflow. Creates one persona per accent (each using a distinct accent voice and mirrori...
Guide for running acceptance tests for a Terraform provider. Use this when asked to run an acceptance test or to run a t...
Check adapter health, latency percentiles, and error rates in real time. Reports against SLA targets.
End-to-end Coval adversarial / red-team testing workflow. Builds one adversarial test set (12 core attack vectors plus l...
Launch and supervise Claude Code or Codex through the CCAM Run API. Use when the user wants to start a monitored agent,...
Configures and runs agents with different adapters including Claude, OpenAI, CrewAI, Lyzr, and GitHub Models. Supports l...
Load this skill when repeated work runs from its `flow.yaml`: stages in order, approvals, gates, run state on disk, inte...
解析中文自然语言中的 AISBench 测评需求，检查 Python 与 AISBench 运行环境，通过分轮追问确认关键配置，将基准名称映射为仓库内的数据集任务，在用户最终确认后生成并执行 AISBench CLI，并返回精度指标和产物路...
Design a small team of AI agents to tackle a complex task in parallel — who does what, how they hand off, and how to kee...
Invoke run_tests via MCP and surface stable / new-failure / flaky / recovered classifications. Use any time you need to...
Run the ReVision server locally and smoke-test it. Use when asked to start, serve, run, or manually verify the ReVision...
Build, install, launch and screenshot a house Android debug variant on an emulator via the Android CLI, and run a journe...
Build, start, and drive the Nuxt Theme Builder in a real browser (Playwright) to verify a change end to end — editor int...
Build and start the Mendix app in Docker, and restart it after changes. Use when asked to run, launch or restart the app...
Trigger phrases: "start trading", "run agent", "start arena agent", "begin trading", "start auto-trading", "stop trading...
End-to-end Coval audio-quality testing workflow. Launches one run per audio-robustness scenario against the same voice a...
Audit dependencies และ security เพื่อตรวจสอบ vulnerabilities
Use for authorized multi-step tasks that should be organized and completed as bounded, verifiable workpacks with fewer m...
Runs the Autoware autonomous-driving stack against CARLA on UE 5.8 over CARLA's native ROS 2 interface — no bridge proce...
Run one planning batch - start an implementation agent per unit in its own worktree, log the times, merge, run the full...
รัน benchmark tests เพื่อวัดประสิทธิภาพและเปรียบเทียบ performance
Measure Lodestone's renderer with its benchmark against the recorded baseline, or deliberately record a new baseline. Us...
Run and interpret the File API vs Inline benchmark for Gemini performance testing. Use when discussing performance optim...
รัน build process อย่างเป็นระบบเพื่อสร้าง production-ready artifacts
Runs an end-to-end marketing campaign — sales analysis, content brief, Canva assets, HubSpot send. Accepts optional look...
Use when designing or evaluating information architecture — navigation structure, menu labels, content categorization —...
Starts a CARLA UE 5.8 server from a ue58-dev tree — the editor binary in -game mode (the working headless server), the p...
Build, start, run and drive cc-backend end to end — an isolated server with the internal metric store, ingest line proto...
Build, run, drive, screenshot, and smoke-test the cgagentharness console (Rust axum server + browser UI) on a headless L...
รัน lint, typecheck และ scan เพื่อตรวจสอบคุณภาพ
Re-run this repository's automated skill-loading checks against the installed harnesses (Antigravity CLI, Claude Code, C...
Run the project's full verification gate: every check the project defines as a pass/fail condition, built from its CI co...
Clean build artifacts และ cache เพื่อเริ่มต้นใหม่
รัน cleanup tasks เพื่อลบ build artifacts และ cache
Start a single Gambit game client that connects to 127.0.0.1:1234. Use when the user wants to run a client, test client...
Next.js 프로젝트 통합 리뷰. review-task 6단계 + semicolon-reviewer 코드 리뷰를하나의 워크플로우로 통합 실행합니다. PR에 리뷰 코멘트를 자동 등록합니다.Use when (1) "/...
Execute comprehensive test suite (Vitest/Jest) with coverage reporting and failure analysis. Returns structured output w...
Run configuration handling for persistent command configuration storage
Launch INAV Configurator in development mode for testing
Run the CPython regression test suite in this repo via scripts/run_cpython_tests.sh with parallel execution, enforce a p...
Run Playwright E2E (tests/e2e/) against the docker/ all-in-one harness so it reflects LOCAL code changes, not a stale ca...
Run reproducible DeepSWE coding-agent benchmark evaluations through OpenRouter and mini-swe-agent.
Project-specific run and verify recipe for the DegenExus CLI/TUI simulator.
Deploy application ไปยัง platform ที่กำหนด พร้อม commit, push และ watch จน live
Use when a product team needs to answer a critical design or business question through rapid prototyping and user testin...
รัน development server และแก้ไขข้อผิดพลาด
Launch the PeakyPanes dev workflow by running scripts/dev-run in the current repo. Use when the user says "run dev" or a...
Issue 作成から PR 作成まで自動実行する統合開発ワークフロー。「開発フロー」「dev-flow」「Issue から PR まで」「自動開発」「ワークフロー実行」「開発を自動化」「フロー実行」などで起動。計画書に基づいた自動実装を実行...
Build, install, and launch an iOS app on a physical iPhone or iPad entirely from the command line (no Xcode GUI), using...
Build, launch, and drive the DevSkyy Next.js dashboard at `frontend/` headlessly. Use when asked to run, start, launch,...
Deep post-mortem digging into a weft execution. Dispatched with an execution id or a symptom when the cause is unclear o...
รัน documentation site และแก้ไขข้อผิดพลาด