Top Agent Skills

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

Use this skill when the user requests to generate, create, imagine, or visualize images including characters, scenes, products, or any visual content. Supports structured prompts and reference images for guided generation.

Set up GitHub authentication for the agent using git (universally available) or the gh CLI. Covers HTTPS tokens, SSH keys, credential helpers, and gh auth — with a detection flow to pick the right method automatically.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Production pipeline for ASCII art video — any format. Converts video/audio/images/generative input into colored ASCII character video output (MP4, GIF, image sequence). Covers: video-to-ASCII conversion, audio-reactive music visualizers, generative ASCII art animations, hybrid video+audio reactive, text/lyrics overlays, real-time terminal rendering. Use when users request: ASCII video, text art video, terminal-style video, character art animation, retro text visualization, audio visualizer in ASCII, converting video to ASCII art, matrix-style effects, or any animated ASCII output.

Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization

Deploy applications and websites to Vercel. Use this skill when the user requests deployment actions such as "Deploy my app", "Deploy this to production", "Create a preview deployment", "Deploy and give me the link", or "Push this live". No authentication required - returns preview URL and claimable deployment link.

Pull data from a web page. First call on a new intent prototypes the flow via $B primitives and returns JSON. Subsequent calls on a matching intent route to a codified browser-skill and return in ~200ms. Read-only — for mutating flows (form fills, clicks, submissions), use /automate. Use when asked to "scrape", "get data from", "pull", "extract from", or "what's on" a page. (gstack)

Self-tuning question sensitivity + developer psychographic for gstack (v1: observational). Review which AskUserQuestion prompts fire across gstack skills, set per-question preferences (never-ask / always-ask / ask-only-for-one-way), inspect the dual-track profile (what you declared vs what your behavior suggests), and enable/disable question tuning. Conversational interface — no CLI syntax required. Use when asked to "tune questions", "stop asking me that", "too many questions", "show my profile", "what questions have I been asked", "show my vibe", "developer profile", or "turn off question tuning". (gstack) Proactively suggest when the user says the same gstack question has come up before, or when they explicitly override a recommendation for the Nth time.

Fix failing or flaky Playwright tests. Use when user says "fix test", "flaky test", "test failing", "debug test", "test broken", "test passes sometimes", or "intermittent failure".

Universal Reasoning Framework implementing λο.τ calculus over holarchic structures. Provides severity-based routing (R0-R3 pipelines), modular cognitive architecture (DEC, EVL, PAT, SYN, MEA, HYP, INT), fractal execution patterns, multi-level validation (η≥4, KROG), and adaptive learning. Triggers on: (1) complex multi-step reasoning, (2) high-stakes decisions requiring validation, (3) research synthesis across domains, (4) system design and architecture, (5) crisis management, (6) performance optimization. Implements scale-invariant reasoning from micro (tool calls) through meso (skill composition) to macro (orchestrated workflows).

Set up Playwright in a project. Use when user says "set up playwright", "add e2e tests", "configure playwright", "testing setup", "init playwright", or "add test infrastructure".

Comprehensive spreadsheet creation, editing, and analysis with support for formulas, formatting, data analysis, and visualization. When Claude needs to work with spreadsheets (.xlsx, .xlsm, .csv, .tsv, etc) for: (1) Creating new spreadsheets with formulas and formatting, (2) Reading or analyzing data, (3) Modify existing spreadsheets while preserving formulas, (4) Data analysis and visualization in spreadsheets, or (5) Recalculating formulas

Personal working rules, decision frameworks, and preferences for ejarbe. Invoke with /rules to surface these during any session. AUTO-TRIGGER when working with MCP servers, MCP transport, SSE, streamable HTTP, server migration, .well-known metadata, or MCP server discovery/registration.

Use when the user asks to inspect Sentry issues or events, summarize recent production errors, or pull basic Sentry health data via the Sentry API; perform read-only queries with the bundled script and require `SENTRY_AUTH_TOKEN`.

Use when the user asks to generate or edit images via the OpenAI Image API (for example: generate image, edit/inpaint/mask, background removal or replacement, transparent background, product shots, concept art, covers, or batch variants); run the bundled CLI (`scripts/image_gen.py`) and require `OPENAI_API_KEY` for live calls.

Generate syntactically correct Claude Code configuration components (commands, subagents, skills, output styles) with optimal cross-component integration. Triggers on requests to create slash commands, agents, subagents, skills, output styles, Claude Code plugins, agentic workflows, or multi-agent pipelines. Ensures parameter validity per component type and semantic coherence across component references.

Prompt engineering standards and context engineering principles for AI agents based on Anthropic best practices. Covers clarity, structure, progressive discovery, and optimization for signal-to-noise ratio.

Metasuperhypergraph orchestrator with fractal scale-invariant recursion. Activates maximum performance mode through renormalization-group optimization, power-law resource allocation, and homoiconic self-improvement. Triggers on "ultrawork", complex multi-domain tasks, or when parallelization/delegation yields superlinear returns.

A runbook written in a wiki decays silently: the dashboard is renamed, the

Serverless is a billing-and-scaling model, not a badge of modernity.

Use this skill for image/video metadata, alt strategy, thumbnails, transcripts, media indexing, performance. Trigger when the task involves seo work related to Image Video SEO, implementation, audits, debugging, strategy, or validation.

Use this skill for navigation, information architecture, menus, tabs, breadcrumbs, repeated workflows. Trigger when the task involves ui/ux work related to Navigation Ia UX, production implementation, audits, debugging, strategy, or validation.

Use this skill for turning issues into implementation plans, branches, code changes, tests, and PR descriptions. Trigger when the task involves programming work related to Issue To PR Agent, implementation, audits, debugging, strategy, or validation.

Use this skill for README, ADRs, runbooks, changelogs, decision records, operator docs. Trigger when the task involves programming work related to Documentation Adrs, production implementation, audits, debugging, strategy, or validation.

Use this skill for scored UI audit, spacing, alignment, states, contrast, hierarchy, polish checklist. Trigger when the task involves web design work related to Design Qa Scorecard, implementation, audits, debugging, strategy, or validation.

Use this skill for test strategy, E2E automation, visual regression, flaky test control, QA evidence. Trigger when the task involves agent profile work related to Qa Automation Agent, implementation, audits, debugging, strategy, or validation.

Use this skill for REST/RPC contracts, versioning, error shapes, schemas, backward compatibility. Trigger when the task involves programming work related to API Interface Design, production implementation, audits, debugging, strategy, or validation.

Use for PPC work when Codex should validate Meta catalog, feed, event, availability, and dynamic product ad readiness.

Agent profile for evaluate RAG quality, chunking, retrieval, citations, hallucination risks, freshness, and regression sets. Use when Codex needs a specialist agent perspective for planning, implementation, review, debugging, validation, or handoff in this domain.

Use this skill for WCAG AA, keyboard access, semantic HTML, focus states, labels, contrast, screen readers. Trigger when the task involves web design work related to Accessibility WCAG Aa, production implementation, audits, debugging, strategy, or validation.

Use this skill for Next.js App Router metadata, robots.ts, sitemap.ts, OG images, dynamic route SEO. Trigger when the task involves seo work related to Nextjs App Router SEO, implementation, audits, debugging, strategy, or validation.

Use this skill for release readiness, rollback plans, changelogs, deployment checks, production risk. Trigger when the task involves agent profile work related to Release Manager Agent, implementation, audits, debugging, strategy, or validation.

Use this skill for WebXR, AR previews, model-viewer, mobile sensors, permissions, fallback UX. Trigger when the task involves 3d web work related to XR AR Web Experiences, production implementation, audits, debugging, strategy, or validation.

Use this skill for queues, retries, idempotency, cron, dead-letter queues, background processing. Trigger when the task involves programming work related to Background Jobs Queues, implementation, audits, debugging, strategy, or validation.

Use this skill for DeFi protocols, TVL, yield sources, contract risk, liquidity, oracle and governance risks. Trigger when the task involves crypto work related to Defi Protocol Analysis, production implementation, audits, debugging, strategy, or validation.

Use for PPC work when Codex should audit Meta Ads account structure, delivery, learning, attribution, creative, audience, and spend evidence.

Use this skill for SaaS UI polish, spacing, density, state design, professional visual hierarchy. Trigger when the task involves ui/ux work related to Premium Saas UI Polish, production implementation, audits, debugging, strategy, or validation.

Use this skill for large sitemap indexes, sharding, freshness, lastmod hygiene, monitoring, submission strategy. Trigger when the task involves seo work related to Sitemap Index Sharding, implementation, audits, debugging, strategy, or validation.

KOMODO v1.0 — Momentum Event Consensus. Uses leaderboard_get_momentum_events (real-time threshold crossings) to detect when 2+ quality SM traders cross momentum thresholds on the same asset/direction within 60 minutes. Confirmed by market concentration + volume. Enters with the momentum. Replaces MANTIS v1.0 and SCORPION v1.1 (both used stale position data).

CHEETAH v2.0 — HYPE Predator. SM commitment as primary signal. Top performer at +7.6%. All fleet fixes applied.

Interact with Hydrex liquidity pools on Base. Use when the user wants to lock HYDX for voting power, check voting power for gauge voting, vote on liquidity pool strategies, view pool information, check voting weights, participate in Hydrex governance, deposit single-sided liquidity into auto-managed vaults to earn Hydrex yields, claim oHYDX rewards from incentive campaigns, or exercise oHYDX into veHYDX. Uses Bankr for transaction execution.

Deploy ERC20 tokens on Base, Ethereum, Arbitrum, and other EVM chains using the Clanker SDK. Use when the user wants to deploy a new token, create a memecoin, set up token vesting, configure airdrops, manage token rewards, claim LP fees, or update token metadata. Supports V4 deployment with vaults, airdrops, dev buys, custom market caps, vanity addresses, and multi-chain deployment.

Accesses Quotient market intelligence through either x402 micropayments or API key auth, with explicit 402 challenge/settle handling when using x402.

Lightweight scanner tracking Smart Money market concentration across all Hyperliquid assets. Flags assets accelerating up the ranks before they become crowded top-3 plays. IMMEDIATE_MOVER signal fires on 10+ rank jumps with quality filters (erratic history, velocity gate, trader count floor, max leverage check). One API call per scan, runs every 60 seconds. Use when detecting SM rotations, finding emerging opportunities early, or monitoring rank acceleration patterns.

This skill should be used when building trading bots, implementing swap functionality, or adding cross-chain bridge support. Covers same-chain swaps, cross-chain bridges, ETH/WETH conversions, and amount formats.

This skill should be used when the user asks about Hyperliquid whale positions, Hyperliquid liquidations, Hyperliquid open interest, Hyperliquid trader analytics, Hyperliquid taker data, smart money on Hyperliquid, or any Hyperliquid-specific query. Use when user says: 'Hyperliquid whales', 'HL whale positions', 'HL liquidations', 'HL open interest', 'HL trader', 'smart money', 'Hyperliquid大户', 'HL鲸鱼', 'HL持仓', 'HL清算', 'HL持仓量', 'HL交易员'. For general crypto prices/news, use aicoin-market. For exchange trading, use aicoin-trading. For Freqtrade, use aicoin-freqtrade.

Call Cheshire Terminal REST, MCP, discovery, and developer API surfaces for Solana agents, arena rooms, Upstash boxes, trading health, and OpenAPI. Use when integrating cheshireterminal.ai, /mcp, /.well-known/agent-card.json, ct_sk_ API keys, arena coordination, box handoff, or Apigee/zero-service health probes.

Set up the MoonPay CLI, authenticate, and manage local wallets. Use when commands fail, for login, or to create/import wallets.

从内容到成品PPTX的端到端演示文稿制作，含AI插画生成和18种设计风格。当用户提到"做PPT"、"做幻灯片"、"演示文稿"、"Keynote"、"slides"时使用。

Build a visual sequence that proves, explains, and supports decisions.

Skeptical pushback on the research dossier before it feeds the outline. Asks whether claims are cited, whether surprising findings are actually surprising, whether we missed the strongest competitor angle. One revision pass on FAIL (BLOG_AGENT_RESEARCH_REVISION_BUDGET, default 1).

Track performance across sessions and reduce scaffolding as competence grows. Makes fading visible — the learner knows when scaffolds are removed and why. Use for sustained learning engagement where independence is the goal.

Audits an existing lesson against UDL's three principles — engagement, representation, and action/expression. Identifies specific access barriers and suggests concrete modifications ranked by impact.

Generates multiple means of engagement, representation, and action/expression for a given learning goal. Produces specific, practical alternatives — not generic options — and recommends the highest-impact single change.

Design a diagnostic hinge question that reveals whether students understand enough to move on. Use when planning key checkpoints mid-lesson during explicit or direct instruction.

Coordinates UDL and differentiation tools through a universal-first hierarchy: barrier removal before targeted differentiation before individualised accommodation. Use when planning accessible learning.

Create an interactive digital worked example sequence with fading for online or blended delivery. Use when building e-learning modules, LMS content, or app-based instruction.

Structure a decision or design challenge through multiple perspectives before committing to action. Use as a synthesis step after scoping, mapping, and dilemma navigation when a group needs a wiser next step.