Profile and optimize Python code using cProfile, memory profilers, and performance best practices. Use when debugging slow Python code, optimizing bottlenecks, or improving application performance.

Adapt and debug existing or new models for vLLM on Ascend NPU. Implement in /vllm-workspace/vllm and /vllm-workspace/vllm-ascend, validate via direct vllm serve from /workspace, and deliver one signed commit in the current repo.

Game design principles. GDD structure, balancing, player psychology,

Kotlin: coroutines, data classes, sealed classes, extension functions, Gradle, KMM multiplatform

AI-powered JavaScript reverse engineering tool. 资深JavaScript逆向工程专家助手。Actions: collect, search, deobfuscate, understand, summarize, detect-crypto, browser, debugger, breakpoint, debug-step, debug-eval, debug-vars, script, hook, stealth, dom, page. Capabilities: 混淆代码分析、VM破解、Webpack解包、AST转换、Puppeteer/CDP自动化、反检测、指纹伪造、加密识别、参数提取、算法还原、Canvas/WebGL指纹、WebDriver隐藏、CDP调试、断点分析、动态追踪、Hook注入、DOM检查、页面控制。

Use when implementing complex multi-step tasks, fixing critical bugs, or when quality and completeness matter more than speed - ensures comprehensive implementation without shortcuts through systematic analysis, implementation, and verification phases

Use when generating static map images with the Python `staticmap` library — embedding isochrone or drive-time polygon maps in Word (.docx) reports, placing facility or location markers over polygon fills using PIL, sizing maps for page layout, or debugging CircleMarker visibility, `m.transformer` AttributeErrors, or coordinate conversion failures. Triggers on: trade area map, isochrone map, drive-time map, map image too large, legend pushed to next page, markers not appearing, blank page in docx.

Fix a bug following the Reproduce > Diagnose > Fix > Verify workflow. Use when something is broken, there's an error, or the user reports unexpected behavior.

Verify AWS infrastructure configuration before deployment. Use when validating VPC endpoints, NAT Gateway capacity, security groups, or debugging network path issues that cause Lambda connection timeouts.

Guides development and maintenance of the FigUI3 web components library for Figma-style plugin UIs. Applies when adding or modifying `fig-*` custom elements, updating docs/demo pages, adjusting theme tokens, improving accessibility, or debugging component behavior in `fig.js`, `components.css`, `index.html`, and `README.md`.

Debug workflow per foundation command.

Investigate dependency source code and manage external repos via repos.yaml. Use when investigating how a library works internally, exploring dependency source, reading upstream code, debugging third-party behavior, adding a new dependency to track, contributing to upstream forks, or when the user mentions repos.yaml, repo manifest, or upstream repos. Preferred over searching node_modules.

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.

Guidance for `effect/Console` focused on APIs like dir, log, and info. Load after `effect-skill-router` when this module is the primary owner.

Macro and procedural metaprogramming expert covering macro_rules!, derive macros, proc-macros, compile-time computation, and code generation patterns.

Tune and debug World42 terrain shadow mapping across near/far directional cascades, TerrainShadowContext wiring, terrain shader sampling, and chunk caster registration. Use when shadows flicker, acne or peter-panning appears, cascade seams are visible, shadows drift during camera motion, or when editing src/app/setup_lod_and_shadows.ts, src/game_objects/planets/rocky_planet/terrains_shader.ts, src/assets/shaders/terrain/terrainFragmentShader.glsl, and src/systems/lod/chunks/chunk_forge.ts.

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.

**Skill Name**: browser-local-chrome-remote-debugging

workleap-telemetry-v2

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.

workleap-telemetry

Use when building or debugging interactive 3D scenes on the web with Three.js (scene/camera/renderer, lights/materials, GLTF loading, controls, performance). Helpful for designers shipping 3D UI moments.

Use when implementing 2D physics interactions with Matter.js, including Engine/World setup, Render/Runner configuration, adding bodies and constraints, and scroll/interaction-friendly canvas scenes.

Structured code debugging through hypothesis formation and falsification planning. Use when diagnosing bugs, unexpected behaviour, or system failures where the root cause is unclear. Produces a hypothesis document for execution by another agent rather than performing the investigation directly. Triggers on requests to debug issues, diagnose problems, investigate failures, or create debugging plans.

Comprehensive guide to implementing RAG systems including vector database selection, chunking strategies, embedding models, and retrieval optimization. Use when building RAG systems, implementing semantic search, optimizing retrieval quality, or debugging RAG performance issues.

Linux kernel development skill grounded in local references/labs and references/lectures materials. Use for kernel modules, system calls, process scheduling, interrupts, locking, memory management, filesystems, networking, architecture, debugging, profiling, and device model work.

AI behavioral guideline for autonomously detecting work session boundaries and proposing updates to current_focus.md. The AI monitors conversation flow for natural breakpoints and acts without explicit invocation.

End-to-end MegaETH development playbook (Feb 2026). Covers wallet operations, token swaps (Kyber Network), eth_sendRawTransactionSync (EIP-7966) for instant receipts, JSON-RPC batching, real-time mini-block subscriptions, storage-aware contract patterns (Solady RedBlackTreeLib), MegaEVM gas model, WebSocket keepalive, bridging from Ethereum, and debugging with mega-evme. Use when building on MegaETH, managing wallets, sending transactions, or deploying contracts.

Complete Claude Code productivity system — project setup, prompting patterns, sub-agent orchestration, context management, debugging, refactoring, TDD, and shipping 10X faster. Zero scripts needed.

Build, integrate, secure, and scale MCP servers and clients. From first server to production multi-tool architecture.

You are an expert n8n workflow architect. You design, build, debug, optimize, and scale n8n automations following production-grade methodology. Every workflow you create is complete, functional, and f

> Build production RAG systems that actually work. From chunking strategy to evaluation — the complete methodology.

Help develop, build, flash, and debug ESP32/ESP8266 firmware using Espressif ESP-IDF on Linux/WSL. Use when the user asks about ESP-IDF project setup, configuring targets, menuconfig, building, flashing via esptool/idf.py, serial monitor, partition tables, sdkconfig, troubleshooting build/flash/monitor errors, or automating common idf.py workflows from the command line.

Diagnose and fix common frontend issues — white screen, JS errors, resource loading failures, React/Vue hydration, browser extension popup, and CSS layout bugs.

Systematic approach to debugging and problem-solving. Distilled from real production incidents and industry best practices.

Track skill execution details including matched skills, decomposed tasks, execution status, outputs, and timestamps. Called at the start of every conversation turn to record skill usage.

Install and verify the SafeClaw safety proxy for OpenClaw and other OpenAI-compatible clients. Use when setting up a local or hosted SafeClaw proxy, routing model traffic through it, debugging why calls are not appearing in the dashboard, or adapting setup for OpenClaw provider configs, agent models.json, or shell-launched clients.

Control Slack via Browser Automation to send messages, manage huddles, screen share, set status, and react as the logged-in user.

Lybic Sandbox is a cloud sandbox built for agents and automation workflows. Think of it as a disposable cloud computer you can spin up on demand. Agents can perform GUI actions like seeing the screen, clicking, typing, and handling pop ups, which makes it a great fit for legacy apps and complex flows where APIs are missing or incomplete. It is designed for control and observability. You can monitor execution in real time, stop it when needed, and use logs and replay to debug, reproduce runs, and evaluate reliability. For long running tasks, iterative experimentation, or sensitive environments, sandboxed execution helps reduce risk and operational overhead.

Provides NodeReal MegaNode blockchain infrastructure APIs for 25+ chains including BSC, Ethereum, opBNB, Optimism, Polygon, Arbitrum, and Klaytn. Covers standard JSON-RPC endpoints, Enhanced APIs (nr_ methods for ERC-20 token balances, NFT holdings, asset transfers), MegaFuel gasless transactions via BEP-322 paymaster, Direct Route MEV protection, Debug/Trace APIs, WebSocket subscriptions, ETH Beacon Chain consensus layer, Portal API usage monitoring, API Marketplace (NFTScan, Contracts API, SPACE ID, Greenfield, BNB Staking, PancakeSwap, zkSync), non-EVM chains (Aptos, NEAR, Avalanche), and JWT authentication. Use when building blockchain dApps with NodeReal, querying token or NFT data, setting up RPC infrastructure, configuring gasless transactions, protecting against MEV, tracing transactions, verifying smart contracts, resolving .bnb domains, or monitoring validators and API usage.

Design, configure, debug, and optimize OpenClaw AI agent deployments. Master guide for gateway configuration, openclaw.json settings, model routing and fallback chains, skills development and publishing, cron job scheduling, memory systems (Qdrant, Neo4j, SQLite), Docker infrastructure, and Tailscale VPN networking. Includes config analyzer that audits your openclaw.json and suggests improvements, plus health checker that validates all OpenClaw subsystems. Built for AI agents — Python stdlib only, no dependencies. Use for OpenClaw setup, gateway debugging, skill building, cron management, model optimization, cost reduction, and infrastructure troubleshooting.

Structured reasoning through sequential thinking — break complex problems into steps, solve each independently, verify consistency, synthesize conclusions with confidence scoring. Use for complex analysis, debugging, and multi-step reasoning.

Identifica e resolve problemas comuns em AgentSkills do OpenClaw, incluindo: (1) Falhas de instalação via ClawHub, (2) Inconsistências de configuração (ex: paths absolutos), (3) Dependências ausentes ou mal documentadas, e (4) Erros de execução de scripts ou integração com APIs. Fornece um guia passo a passo para depurar e validar skills, garantindo sua correta operação em ambientes de quarentena ou produção. Use esta skill quando uma AgentSkill não se comporta como esperado ou você precisa auditar seu código.

Set up, audit, and debug analytics tracking implementation — GA4, Google Tag Manager, event taxonomy, conversion tracking, and data quality. Use when building a tracking plan from scratch, auditing existing analytics for gaps or errors, debugging missing events, or setting up GTM. Trigger keywords: GA4 setup, Google Tag Manager, GTM, event tracking, analytics implementation, conversion tracking, tracking plan, event taxonomy, custom dimensions, UTM tracking, analytics audit, missing events, tracking broken. NOT for analyzing marketing campaign data — use campaign-analytics for that. NOT for BI dashboards — use product-analytics for in-product event analysis.

Short alias for external-site-profile-learning. Use this when investigating, adding, validating, or debugging external website profiles for the 99idea Playwright browser demo.

Edge browser automation via Chrome DevTools Protocol (CDP). List tabs, navigate, take screenshots, extract page content/HTML, execute JavaScript, click elements, type text, fill forms, scroll. Use when needing to control Edge browser, scrape web content, automate web forms, or take browser screenshots on Windows. Requires Edge with --remote-debugging-port=9222.

Comprehensive deep reasoning framework that guides systematic, thorough thinking for complex tasks. Automatically applies for multi-step problems, ambiguous requirements, architectural decisions, debugging sessions, and any task requiring careful analysis beyond surface-level responses. Use when the task is complex, has multiple valid approaches, involves trade-offs, or when the user asks to think deeply or carefully.

Compete in TitleClash - write creative titles for images and win votes. Use when user wants to play TitleClash, submit titles, or check competition results.

4-phase root cause debugging. Never guess, find the cause. Based on real production debugging methodology. Trigger on: 'bug', 'error', 'not working', 'broken', 'debug', 'fix this'.

Execute ArcAgent bounty workflows end-to-end via MCP tools. Use when claiming bounties, implementing in workspace, submitting for verification, debugging worker/workspace issues, and iterating failed runs until pass. Continue retry/resubmit loops based on verification feedback until either (1) verified PR and payout success or (2) explicitly giving up and releasing the claim.