Skills（SKILL.md）は、AIエージェント（Claude Code、Cursor、Codexなど）に特定の能力を追加するための設定ファイルです。
詳しく見る →MCP server configuration and management. Use when setting up MCP servers, troubleshooting MCP issues, or adding new integrations. Keywords: MCP, model context protocol, supabase, posthog, playwright, gtm, analytics, mcp.json, environment variables, oauth
Debug MCP server communication. Use for troubleshooting MCP integrations, viewing traffic, and analyzing latency.
Use this skill when you need to discover MCP tool parameters, understand naming conventions, or debug tool call errors
**Core principle:** MCP Qdrant-Indexer tools are primary for exploring code and history. Start with MCP for exploration, debugging, or "where/why" questions; use literal search/file-open only for narr
Systematic troubleshooting for MCP tools including installation issues, runtime errors, config problems, and debugging strategies. Use when MCP tools not working, debugging errors, or helping users...
MediaPipe pose detection expertise. Use when debugging landmark tracking, adjusting confidence thresholds, fixing pose detection issues, working with pose.py and video_io.py, or validating pose detection with manual observation.
Execute and troubleshoot memory-cli commands for episode management, pattern analysis, and storage operations. Use this skill when running CLI commands, debugging CLI issues, explaining command usage, or guiding users through CLI workflows.
Debug memory leaks, profile memory usage, optimize allocations. Use when heap grows unexpectedly, OOM errors occur, or profiling shows memory bottleneck. Covers C++ (Valgrind, ASAN, RAII), Python (tracemalloc, objgraph), and general patterns.
Use and troubleshoot the Memory MCP server for episodic memory retrieval and pattern analysis. Use this skill when working with MCP server tools (query_memory, analyze_patterns, advanced_pattern_analysis), validating the MCP implementation, or debugging MCP server issues.
Instrument RAG retrieval, memory operations, and context management
Guide for writing and improving Claude Code skills. Use when creating new skills, debugging why skills aren't activating, or improving skill descriptions and structure.
Prometheus-compatible metrics collection with counters, gauges, and histograms. Export metrics for dashboards and alerts with proper labeling.
Help craft efficient Mimir/Prometheus queries, troubleshoot metric issues, avoid high-cardinality problems, and recommend best practices for aggregation, recording rules, and performance.
Capture Minecraft Bedrock network packets using a proxy relay. Start packet capture, save raw binary dumps and processed JSON logs. Use when debugging Bedrock protocol, analyzing bot behavior, or recording gameplay packets.
HOUSTON executes feature tasks directly. No agents. Best for small features (1-3 tasks).
Guide for configuring Mixins in Architectury multi-loader projects
Interpret machine learning models using SHAP, LIME, feature importance, partial dependence, and attention visualization for explainability
Diagnose and stabilize ML training runs, recover from failures, and deliver validated fixes.
MLflow 3 GenAI evaluation for agent development. Use when (1) writing mlflow.genai.evaluate() code, (2) creating @scorer functions, (3) building evaluation datasets from traces, (4) using built-in scorers (Guidelines, Correctness, Safety, RetrievalGroundedness), (5) analyzing traces for latency/errors/architecture, (6) optimizing agent context/prompts/token usage, (7) debugging evaluation failures. Covers the full eval workflow: trace analysis -> dataset building -> scorer creation -> evaluation execution.
Enterprise practical workflows, context engineering strategies, JIT (Just-In-Time) retrieval optimization, real-world execution examples, debugging patterns, and moai-adk workflow mastery; activates for workflow pattern learning, context optimization, debugging issue resolution, feature implementation end-to-end, and team knowledge transfer
**Version**: 1.0.0
Guidance for `effect/Console` focused on APIs like dir, log, and info. Load after `effect-skill-router` when this module is the primary owner.
>
Thesys C1 - Generative UI API for AI-native applications. Use for generating dynamic, interactive UI components from natural language prompts, building conversational interfaces, analytics dashboards, and AI agents with rich visual responses.
Diagnose issues in the current Amplifier session — misconfigured tools, failing operations, unexpected behavior. Use when something isn't working right.
workleap-telemetry-v2
workleap-telemetry
Guidance for using @landfolk/tx styling (tx prop, grouping syntax, Tailwind transformer, SWC + ESLint plugins) in React/Next.js.
Diagnose và fix các vấn đề liên quan đến Google ADK agents, tools, session state, multi-agent communication trong dự án MemRAG.
>-
Write and debug TypeQL queries for TypeDB 3.8+. Use when working with TypeDB schemas, data queries, insertions, deletions, or functions.
Schedule and execute background work on iOS using BGTaskScheduler. Use when registering BGAppRefreshTask for short background fetches, BGProcessingTask for long-running maintenance, BGContinuedProcessingTask (iOS 26+) for foreground-started work that continues in background, background URLSession downloads, or background push notifications. Covers Info.plist configuration, expiration handling, task completion, and debugging with simulated launches.
Debug iOS apps and profile performance using LLDB, Memory Graph Debugger, and Instruments. Use when diagnosing crashes, memory leaks, retain cycles, main thread hangs, slow rendering, build failures, or when profiling CPU, memory, energy, and network usage.
Pixel-based motion and UI change analysis from frame sequences or screenshots using computer vision and visual comparison. Use when `.transition` or visual effects are not observable via MotionEyes logs, when you only have screenshots/video, when you need regression diffs across builds, or when you need to summarize on-screen changes without instrumentation. Do not use when MotionEyes traces are available and you need precise timing/values; use `motioneyes-animation-debug` instead.
Build responsive, mobile-first web applications. Use when implementing responsive layouts, touch interactions, mobile navigation, or optimizing for various screen sizes. Triggers on responsive design, mobile-first, breakpoints, touch events, viewport.
Implements declarative Flutter routing using GoRouter v17 with URL-based navigation, deep linking, and authentication guards. Activates when configuring GoRoute path patterns, setting up StatefulShellRoute for persistent bottom navigation state, implementing ShellRoute for shared nested layouts, adding redirect guards with loop prevention (idempotent redirects), configuring deep links for iOS (FlutterDeepLinkingEnabled) or Android (intent-filters), debugging redirect loops or navigation stack issues, handling 404 error routes with errorBuilder, using named routes with path/query/extra parameters, or migrating from imperative Navigator.push to declarative routing. Ideal for Flutter web apps requiring URL bar synchronization, mobile apps with universal links or app links, or multi-level navigation hierarchies.
Background processing and heavy computations using Dart Isolates with Isolate.run(), compute() function, and platform channels. Use this skill when performing CPU-intensive operations without blocking UI (image processing, data parsing, encryption), implementing background tasks that prevent jank or frame drops, handling large file operations (JSON parsing, database migrations), running parallel computations, fixing UI freezes caused by main thread blocking, using compute() for one-off background work, creating long-lived Isolates with ReceivePort/SendPort, implementing multi-threaded algorithms, debugging isolate communication issues, or preventing memory leaks from unclosed ReceivePorts. Covers isolate spawn, message passing, ReceivePort cleanup (CRITICAL for leak prevention), SendPort communication patterns, error handling, serialization rules, and performance optimization for compute-heavy operations.
Provider (v6.x) state management for Flutter using ChangeNotifier and InheritedWidget with MVVM architecture pattern. Use this skill when implementing Provider-based state management, using Consumer/Consumer2 widgets for targeted rebuilds, accessing state with context.read/context.watch/context.select, setting up ChangeNotifierProvider or StateNotifierProvider, configuring MultiProvider for multiple providers, implementing ProxyProvider for dependent state, using Selector for optimized rebuilds, migrating from StatefulWidget to reactive state, maintaining legacy Provider codebases, debugging provider not found errors, preventing memory leaks with proper ChangeNotifier disposal, fixing context access issues in initState, or resolving .value constructor anti-patterns. Essential for apps with shared state across widgets (shopping carts, authentication, global settings, form state, theme management), when teaching state management fundamentals, or when debugging memory leaks and resource cleanup in Provider-based applications.
Assists with debugging TypeScript and JavaScript applications.
Use when implementing or debugging ANY network request, API call, or data fetching. Covers fetch API, axios, React Query, SWR, error handling, caching strategies, offline support.
Provides React Native performance optimization guidelines for FPS, TTI, bundle size, memory leaks, re-renders, and animations. Applies to tasks involving Hermes optimization, JS thread blocking, bridge overhead, FlashList, native modules, or debugging jank and frame drops.
Lightweight research-plan-implement workflow for small tweaks, isolated bugfixes, copy edits, and low-blast-radius polish in this repo. Use when the change can be understood from 1-3 files and stays inside one subsystem without changing shared data flow.
VacuumTube（/opt/VacuumTube/vacuumtube）を Chromium や playwright ではなく直接 remote debugging（CDP, 既定: http://127.0.0.1:9992）で操作して、YouTube TV のホームや視聴画面から BGM・音楽・ニュース動画を選択/再生する。ユーザーが「BGM再生して」「音楽再生して」「ライブニュース再生して」「国連総会の最新ニュース動画見たい」など依頼したときに使う。
Help users install, configure, and operate OpenClaw (gateway, channels, nodes, plugins). Use when answering OpenClaw setup/debug questions; use the local docs snapshot bundled with this skill as the source of truth. Triggers: openclaw, clawdbot, clawd, clawdhub, gateway, onboard, channels login, whatsapp, telegram, discord, mattermost, pairing, nodes, sandboxing, tailscale.
Investigate Claude Tabs debug captures. Use only when the user is explicitly working with debug logs, session IDs, TAP output, proxy traffic, observability files, or debug markers and wants log-based analysis or correlation.
Debug and optimize Android/Gradle build performance. Use when builds are slow, investigating CI/CD performance, analyzing build scans, or identifying compilation bottlenecks.
Error handling patterns for n8n workflows including error workflows, Error Trigger, Stop And Error, Continue On Fail, retry logic, and production error monitoring. Use when implementing error handling, building error workflows, debugging failed executions, or setting up alerting.
Generate structured handoff documents to transfer full context between AI chat sessions. Use when: (1) user says 'compact', 'checkpoint', 'save progress', 'handoff', (2) context is getting long or slow, (3) user wants to resume from a previous handoff with 'resume', 'continue where I left off', 'load handoff'. Proactively suggest after substantial work (5+ file edits, complex debugging, architecture decisions).
Build HTML email templates that render everywhere. Use when designing email layouts, fixing Outlook rendering, implementing dark mode, adding accessibility, or choosing a templating framework.
Maintain email conversation context across messages using threading headers. Use when building thread reconstruction, linking replies to conversations, detecting thread hijacking, stripping quoted content, or providing thread context to AI agents.