Claude Skills by yanacuti1121

Use Venice as a pay-per-call JSON-RPC proxy to 20+ EVM and Starknet networks. Covers GET /crypto/rpc/networks, POST /crypto/rpc/{network}, the 1×/2×/4× method-tier pricing model, per-minute + 24-hour credit rate limits, idempotency keys for safe retries, single vs batch requests, and the unsupported

Call POST /embeddings on Venice. Covers request shape (input, model, encoding_format, dimensions, user), OpenAI compatibility, response compression (gzip/br), and practical usage for retrieval, clustering, and RAG.

Handle Venice API errors correctly. Covers the StandardError / DetailedError / ContentViolationError / X402InferencePaymentRequired body shapes, every meaningful status code (400, 401, 402, 403, 415, 422, 429, 500, 503, 504), the 402 PAYMENT-REQUIRED header used by x402 inference, 422 content-policy

Transform existing images with Venice. Covers POST /image/edit (prompt-driven single-image edit), /image/multi-edit (compose 1-3 images), /image/upscale (2-4x upscale + enhance), and /image/background-remove. Accepts base64, file upload, or HTTPS URL.

Generate images with Venice. Covers POST /image/generate (Venice-native), POST /images/generations (OpenAI-compatible), GET /image/styles (style presets), request fields (prompt, dimensions, cfg_scale, seed, variants, style_preset, aspect_ratio, resolution, safe_mode, watermark), and response format

Discover Venice models, their capabilities, constraints, and pricing. Covers GET /models (with ?type filter), /models/traits, /models/compatibility_mapping, the ModelResponse schema (capabilities, constraints, pricing per type), and how to use this to pick the right model programmatically.

Use Venice's Alpha POST /responses endpoint - an OpenAI-compatible Responses API with typed output blocks (reasoning, message, function_call, web_search_call). Covers request shape, streaming, differences from /chat/completions, supported venice_parameters subset, and E2EE behavior.

Generate and transcribe videos via Venice. Covers the async /video/quote + /video/queue + /video/retrieve + /video/complete loop, text-to-video, image-to-video, video-to-video (upscale), audio input, reference images, scene and element support, plus /video/transcriptions for YouTube URLs.

Manage Venice x402 wallet credits. Covers POST /x402/top-up (payment discovery + signed USDC settlement), GET /x402/balance/{walletAddress}, GET /x402/transactions/{walletAddress}, USDC on Base (chain 8453), minimum $5 top-up, transaction types TOP_UP/CHARGE/REFUND, and the x402 v2 PAYMENT-REQUIRED

Use when building AI-powered web apps with Next.js, React, or Node.js — streaming chat, tool calling, structured output, multi-modal. Triggers on: 'vercel ai', 'ai sdk', 'streaming chat', 'useChat', 'generateText', 'streamText', 'ai chatbot nextjs', 'LLM in React', 'ai sdk tool use', 'provider agnostic LLM'.

통합 검증 엔진 - 서브에이전트 기반 fresh-context 검증 루프 (v6)

Use before claiming any task is complete, fixed, passing, or ready. Triggers on: 'done', 'fixed', 'tests pass', 'should work now', 'looks good', 'complete', 'finished', 'ready to merge', or any expression of satisfaction before showing evidence.

프로젝트의 모든 verify 스킬을 실행하여 통합 패턴 검증 보고서를 생성합니다. 기능 구현 후, PR 전, 코드 리뷰 시 사용.

End-to-end production QA, build verification, and launch-readiness checklist for fullstack Next.js apps before going live or shipping a major update. Covers TypeScript, linting, tests, build, SEO tags, route regression, and sitemap validation.

Video translation, dubbing and subtitle generation pipeline — WhisperX speech recognition, LLM translation (Claude/GPT/DeepSeek), multi-voice TTS, Netflix-standard single-line subtitles. Triggers on: 'video translation', 'dịch video', 'thuyết minh video', 'video dubbing', 'tạo phụ đề', 'subtitle generation', 'videolinggo', 'video linggo', 'dịch phụ đề youtube', 'whisperx subtitle', 'tts dubbing video', 'lồng tiếng video AI', 'translate youtube video'.

Translate visa application documents (images) to English and create a bilingual PDF with original and translation

Vite build tool patterns including config, plugins, HMR, env variables, proxy setup, SSR, library mode, dependency pre-bundling, and build optimization. Activate when working with vite.config.ts, Vite plugins, or Vite-based projects.

vLLM PagedAttention KV-cache management for high-throughput LLM serving. Continuous batching, memory utilization, OpenAI-compatible API, and multi-LoRA serving patterns. Sources: vllm-project/vllm (Apache-2.0).

Tích hợp cổng thanh toán VNPay vào ứng dụng Node.js/TypeScript. Dùng khi cần tạo URL thanh toán, xác thực callback, hoặc tích hợp VNPay vào backend. Triggers: "vnpay", "thanh toán vnpay", "tích hợp vnpay", "vnpay payment", "cổng thanh toán", "payment gateway vietnam", "vnpay callback", "vnpay return url", "nestjs-vnpay"

Use when building a voxel (cube-based) world engine from first principles — chunking, mesh generation, terrain generation. Not for general 3D rendering or using an existing voxel engine. Triggers on: 'build a voxel engine', 'minecraft-style world generation', 'chunk meshing', 'greedy meshing algorithm', 'voxel terrain generation', 'implement a block world'. Covers chunk representation, face culling, greedy meshing, and noise-based terrain.

Three-phase vulnerability chain analysis — parallel agents find individual weaknesses, then a synthesis step identifies which combinations escalate to critical impact. Inspired by Strix (usestrix/strix) "Graph of Agents" pentesting model.

WAMP (Web Application Messaging Protocol) secure Pub/Sub for distributed agent event bus. Autobahn|JS patterns, authenticated subscriptions, RPC over WAMP, topic-based routing, and cryptographic message signing. Sources: crossbario/autobahn-js.

wasm-bindgen Rust↔JavaScript interop — export Rust functions/structs to JS, import JS APIs into Rust, wasm-pack build pipeline, TypeScript type generation, memory management across the boundary. Sources: rustwasm/wasm-bindgen (MIT/Apache-2.0).

Wasmer universal WASM runtime — Singlepass/Cranelift/LLVM backend selection, WASI support, embedded runtimes in Rust/Go/JS/Python/C. Run WASM modules from any host language. Sources: wasmerio/wasmer (MIT).

Wasmtime WASI sandbox for agent-executed code — capability-based security, component model, WASI Preview 2. Isolate untrusted code from the host filesystem/network. Sources: bytecodealliance/wasmtime (Apache-2.0).

Audit and optimize web performance — Core Web Vitals (LCP, INP, CLS), bundle splitting, lazy loading, image optimization, caching strategies, and resource hints. Use when asked to "improve performance", "fix LCP/CLS/INP", "optimize bundle", "reduce load time", "lazy load images", "add caching", or before marking any public-facing page as production-ready. Do NOT use for server-side database performance — this covers frontend and network-layer only.

Design and implement WebSocket connections — connection lifecycle, message protocol, authentication, reconnection strategy, backpressure, and horizontal scaling with a pub/sub broker. Use when asked to "add WebSocket", "real-time updates", "live feed", "Socket.IO", "ws library", "reconnect logic", "backpressure", "scale WebSocket", or "pub/sub over WebSocket". Do NOT use for: one-way server push with no client messages — use SSE instead. Do NOT use for: request/response APIs where HTTP pollin...

Low-level WebSocket protocol implementation patterns and code generation for network data structures. Frame parsing, ping/pong keepalive, binary protocol codegen from schema, and bandwidth-optimized message design. Sources: warmcat/libwebsockets (MIT).

E2E testing for Windows native desktop apps (WPF, WinForms, Win32/MFC, Qt) using pywinauto and Windows UI Automation.

Use when provisioning a headless Windows 11 VM in Docker for testing, running Windows-only tools, or installing Claude Code on Windows. Triggers on: 'windows vm', 'windows docker', 'headless windows', 'windows 11 container', 'windows vm ssh', 'kvm windows', 'dockur windows', 'windows dev environment', 'windows claude code', 'windows test environment'.

Use when starting a new feature task, before making experimental changes, or when user says 'try this', 'experiment with', 'test out'. Ensures work happens on a branch, not directly on main.

Write implementation plans that make coding obvious — exact file paths, full code examples, step-by-step commands, expected output. Use before any feature with 3+ files or when delegating to subagents.

Use when creating a new YAMTAM skill. Triggers on: 'add a skill', 'create a skill', 'write a skill', 'new skill for X', 'make a skill that'. Enforces frontmatter shape, trigger-phrase quality, and lockfile registration.

X/Twitter API integration for posting tweets, threads, reading timelines, search, and analytics. Covers OAuth auth patterns, rate limits, and platform-native content posting. Use when the user wants to interact with X programmatically.

XState finite state machines and actor model for agent lifecycle management. State/transition definitions, guards, actions, spawned child actors, and agent state visualization. Sources: statelyai/xstate (MIT).

Safe YAML parsing resistant to YAML bombs and prototype pollution. eemeli/yaml safe schema, document size limits, no-execute mode, type-safe output, and config file validation patterns. Sources: eemeli/yaml.

Yana's task router — classify a task description into simple/complex/external and decide next action. Use before handling any non-trivial request. Triggers on: 'classify task', 'route task', 'yana classify', 'what kind of task', 'should I handle this myself', 'dispatch or handle', 'yana router', 'task type', 'route request', 'auto or agent'.

Yjs CRDT shared types for conflict-free real-time multi-agent data sync. Y.Map/Y.Array/Y.Text operations, awareness protocol, WebSocket provider, and offline-first merge semantics. Sources: yjs/yjs (MIT).

Implement Zero Trust networking for multi-agent swarms. mTLS mutual authentication, egress proxy whitelisting, DNS-over-HTTPS pinning, Ed25519 message signing, and network segmentation for agent clusters.

Apply zero-trust architecture to services — mTLS with SPIFFE/SPIRE, per-request authentication, short-lived credentials, least-privilege service accounts, network policy enforcement, and audit logging every access decision. Use when asked about "zero trust", "ZTA", "never trust always verify", "mTLS", "SPIFFE", "SPIRE", "service identity", "network policy", "per-request auth", "service mesh security", "lateral movement prevention", "east-west traffic", or "microservice auth". Do NOT use for: ...

In-memory ZIP read/write operations without temp files. adm-zip patterns for extracting, modifying, and inspecting ZIP archives entirely in RAM — ideal for release package verification. Sources: cthackers/adm-zip.

Zero-Knowledge Proof patterns for privacy-preserving agent attestation. snarkjs circuit compilation, proof generation, verification, and agent compliance attestation without revealing sensitive source code. Sources: iden3/snarkjs.

Node.js zlib streaming compression for log rotation and release packaging. Gzip/Brotli/Deflate transform streams, compression level tuning, streaming decompress, and on-the-fly log archiving. Sources: Node.js zlib module, madler/zlib.

Detecting and reaping zombie/orphan processes in agent sandboxes. SIGCHLD handling, waitpid reaping, subreaper patterns, and Node.js child_process cleanup. Sources: node-modules/zombie, Linux waitpid(2) man page.

Human-gated skill self-improvement loop. Runs a skill against test scenarios, diagnoses failures, proposes ONE change at a time, and waits for human approval before applying anything. Usage: /improve-skill path/to/SKILL.md

HexStrike AI — MCP server cho 150+ cybersecurity tools: Nmap, SQLMap, Nuclei, Ghidra, Prowler... AI agent tự chọn tool, chạy pentest, bug bounty automation. Cần authorization.

MiroFish — Swarm Intelligence Engine: upload seed data → simulate thousands of agents → predict future trajectories. GraphRAG + multi-agent simulation + interactive chat.

Set up 9Router as a local AI gateway so coding agents never stop when a provider quota runs out — one OpenAI-compatible endpoint (localhost:20128) fanning out to 40+ providers with automatic fallback. Use when asked to 'set up 9router', 'cài 9router', 'hết quota Claude thì làm sao', 'fallback provider tự động', 'never hit rate limits', 'free AI router', or 'nối Claude Code vào nhiều provider'. Do NOT use for: cloud gateway architecture comparisons (LiteLLM/Portkey/Kong) — see ai-gateway-patte...

Agent-to-Agent (A2A) protocol — Google 2025, 150+ org backing. Agent Cards discovery, task lifecycle (submitted→working→completed), artifacts (text/structured/video), opaque task model. MCP vs A2A split. Auth: bearer/mTLS/signed. Sources: rohitg00/ai-engineering-from-scratch (Apache-2.0).

Audit UI code for WCAG 2.1 AA compliance — color contrast, keyboard navigation, ARIA roles, focus management, and screen reader compatibility. Use when the user asks to "check accessibility", "make this accessible", "WCAG audit", "a11y review", or before marking any public-facing UI as done. Do NOT use for quick pre-delivery checks — use output-enforcement for that. This skill goes deeper: it covers ARIA semantics, focus order, and live regions.