Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Places existing images onto an Excalidraw whiteboard, turns a slide deck into clean per-slide images, and inspects a .excalidraw file. Use whenever someone mentions Excalidraw, a whiteboard/canvas, .excalidraw, 白板 or 画板 — e.g. "put these screenshots on my board" — or a workshop delivered by hand-drawing over screenshots on a shared canvas. Not for generating a diagram from a text description.
Token and address risk assessment. Use this skill ONLY when the user's query is exclusively about token/contract/address security with no other analysis dimensions. Trigger phrases: is this token safe, check contract risk, is this address safe, honeypot, rug. If the query ALSO mentions fundamentals, technicals, news, sentiment, or any other analysis dimension, use gate-info-research instead — it handles multi-dimension queries in a single unified report. Address risk mode (is this address safe) is exclusive to this skill and must NOT be routed to gate-info-research.
Use this skill when creating or redesigning a GitHub profile README and pinned-project strategy so the profile communicates credibility, project taste, current focus, writing, and practical authority without sounding arrogant or exposing private employer work.
When the user wants help with Google Ads keyword research, keyword strategy, match types, search term analysis, keyword planning, keyword organization, or keyword audits. Triggers on 'keyword research', 'keyword strategy', 'match types', 'search terms', 'keyword planner', 'keyword audit', 'broad match', 'phrase match', 'exact match', 'keyword expansion', or 'keyword list'. For deep negative keyword optimization, wasted spend elimination, or cannibalization prevention see google-ads-negative-keywords. For building search campaigns see google-ads-search. For bid management on keywords see google-ads-bidding
When the user wants to define GTM metrics, build a metrics dashboard, measure pipeline efficiency, or track AI product performance. Also use when the user mentions 'GTM metrics,' 'revenue latency,' 'pipeline metrics,' 'TTFV,' 'time-to-first-value,' 'data health,' 'attribution,' 'conversion rate,' 'CAC,' 'LTV,' 'NRR,' 'GTM dashboard,' 'magic number,' 'pipeline velocity,' or 'funnel metrics.' This skill covers GTM measurement from metric selection through dashboard design, including AI-specific cost metrics, attribution models, and weekly review cadences. Do NOT use for technical implementation, code review, or software architecture.
Comprehensive parser and explainer for the HPK format (proprietary healthcare message format). Supports over 100 message types covering patient administration (ID, MV, CV), supply chain (PR, FO, MA, CO, LI, RO, FA), inventory (SO, IM), organizational structure (ST, UT), and financial operations (RD, DD). Uses @erp-pas/hpk-dictionary as the source of truth. Validates structure, extracts fields, explains business context, maps to HL7 v2.5/IHE PAM, and helps troubleshoot integration issues. Use when the user provides or asks to analyze an HPK-format message, identify an HPK message type, extract fields, or explain HPK business context. For standard HL7 v2.5 / IHE PAM messages, use hl7-pam-parser instead
Open an HTML file, Markdown file, or localhost page in the browser so the user can edit text directly and leave comments on specific parts, then send all edits and comments back to you. Use after writing or updating something the user will read — specs, plans, reports, newsletter drafts, landing pages, slide decks, and locally running web pages
Integrates OWASP ZAP (Zed Attack Proxy) into GitHub Actions and GitLab CI pipelines, covering baseline, full, and API scan configuration against running applications, ZAP finding interpretation, scan policy tuning, and DAST quality gates. Use when testing running web apps or REST/GraphQL APIs for XSS, SQLi, CSRF, and auth/authz flaws, or when SAST alone is insufficient and runtime DAST is required for compliance or release gating.
Calculate municipal property tax (arnona) for Israeli properties, check discount eligibility, and draft appeal letters to arnona committees. Use when a user needs to estimate arnona payments by municipality, zone, and property usage type, verify eligibility for discounts (olim, soldiers, elderly, disabled, low income, students, single parents), or prepare formal appeals with legal references. Covers all major Israeli municipalities including Tel Aviv, Jerusalem, Haifa, and Beer Sheva. Do NOT use for income tax (mas hachnasa), VAT (maam), or national insurance (bituach leumi) calculations, which fall under separate Israeli tax authorities.
Build and maintain an LLM-powered personal knowledge base or wiki. Use this skill when the user wants to organize research notes, build a second brain, set up an Obsidian vault with LLM compilation, ingest papers or articles automatically, or mentions "personal knowledge base", "LLM wiki", "second brain", "knowledge management", "ingest papers", "obsidian + LLM". Based on Karpathy Gist and 59k-like post.
Write explicit strategy into system prompts as a learning mechanism — making LLM behavior more predictable and teachable through prompt-as-textbook approach. Use this skill when the user wants to improve an agent or LLM's performance on a specific task through system prompts, needs to encode a strategy or workflow into a prompt, wants consistent LLM behavior, or says "improve system prompt", "encode this strategy", "prompt engineering", "make agent follow this workflow", "prompt as textbook". Based on Karpathy system prompt learning post.
A Douyin keyword search and structured evidence-gathering adapter that uses only the installed and logged-in OpenCLI; answers search requests independently or gathers this episode’s Douyin evidence for 真人口播①, but does not decide the final topic. Use when the user says “抖音搜索,” “搜一下抖音,” “抖音上搜,” “抖音关于,” “抖音上有什么,” or “douyin搜索,” or when the bianpai/topic-selection workflow requires validating Douyin keywords.
Manage Letta AI agent fleets declaratively with a kubectl-style CLI. Use when creating, updating, or managing multiple Letta agents with shared configurations, memory blocks, tools, folders, canary deployments, multi-tenancy, and bulk operations.
Load this skill whenever the project supports light/dark mode, color theme switching, high-contrast mode, or responds to prefers-color-scheme. Under no circumstances hard-code colors that break in alternative themes. Always test color contrast in both light and dark themes, and respect user OS-level color preferences via CSS media queries.
Manage Linear issues, projects, and workflows via CLI. This skill should be used when the user wants to create, list, update, or search Linear issues, manage projects or milestones, or interact with their Linear workspace. Triggers on "create a task", "add a Linear issue", "list my issues", "update GLE-123", "what's in my backlog", or any Linear-related request.
Search and analyze TikTok product data, including sales, influencer sales, pricing, and commission rates, across 16 TikTok Shop sites. Trigger this skill when users mention TikTok product search, TikTok Shop product analysis, TikTok sales data, influencer sales, TikTok product selection, TikTok commission rates, TikTok product rankings, EchoTik data queries, TikTok product search, TikTok sales, influencer sales, TikTok commission, TikTok product selection, short-video e-commerce, or TikTok data. Even if users do not explicitly mention “EchoTik” or “TikTok,” trigger this skill whenever their needs involve searching for products on TikTok Shop or analyzing TikTok product performance metrics.
Use FastMoss data to query best-selling product rankings in the global TikTok e-commerce market, with analysis by day, week, month, and category. Trigger for TikTok bestseller rankings, sales rankings, GMV rankings, category bestsellers, weekly product reports, or FastMoss. Also trigger when users want TikTok best-seller rankings or sales rankings by time period, even if “FastMoss” is not mentioned.
Query historical time-series data for Amazon products, including price trends, BSR (Best Sellers Rank) trends, rating changes, seller counts, and monthly sales. Supports any ASIN across multiple Amazon marketplaces. Trigger for price history or tracking, BSR history or trends, historical pricing, price fluctuations, Keepa data, ranking history, price-drop alerts, historical Lightning Deal prices, Buy Box price trends, coupon prices, FBA/FBM price comparisons, changes in seller counts, rating trends, sales history, or related requests. Also trigger for Amazon product-level historical trends in price, rank, or sales, even if “Keepa” or “time-series data” is not mentioned.
Use SellerSprite data to find and analyze competitors on Amazon across 12 marketplaces, including product metrics such as sales, BSR, pricing, ratings, and growth trends. Trigger when users mention competitor lookup or analysis, ASIN reverse lookup, competitive product research, finding similar products, market competitor discovery, product benchmarking, competitor sales estimates, analyzing competing listings, competitor analysis, ASIN reverse lookup, competitor sales, competitor research, SellerSprite, market competitor discovery, or competitor trends. Also trigger even without explicitly mentioning “SellerSprite” or “competitor lookup” if the request involves discovering and analyzing Amazon competitors using ASINs, keywords, seller names, brands, or categories.
Write, register, and debug MaaFramework Custom recognition/actions — criteria for when to use Custom, host and registration paths, binding implementation and registration, parameter parsing and validation, reusing pipeline nodes, agent child process mode, and engineering conventions. Use when implementing MaaFW custom recognition or action logic, deciding registration path for your host, registering customs, parsing custom_action_param/custom_recognition_param, reusing pipeline nodes inside custom code, or setting up the agent child process mode.
Use when structuring a PlayCanvas game's control flow — a small state machine such as ready, playing, paused, and over, pointer-lock capture, pausing on focus or pointer-lock loss, a full reset, and a stable clock and timestep — so the loop stays deterministic and recoverable.
Shared foundations for building reusable data models in PostHog, on either of two stacks: PostHog-native data-warehouse views / materialized views (HogQL, via the view-* MCP tools), or an external dbt project (sources.yml + staging/marts + schema tests) run against your own or PostHog's managed warehouse. Read before authoring any specific business model — covers the PostHog-vs-dbt decision, the view-create → view-materialize → sync_frequency workflow and the HogQL column-aliasing rule, the dbt project skeleton and the "no native dbt integration" picture, warehouse joins and star-schema dimensions, currency conversion with convertCurrency(), and checking/registering models in the data catalog for reuse. Companion to modeling-revenue-metrics, modeling-conversion-metrics, modeling-activation-metrics, modeling-product-usage-metrics, and modeling-dimension-tables. Use when the user asks how to build a view, materialized view, or dbt model in PostHog, or which of the two stacks to use.
Framer Motion (Motion) animation patterns for React — springs, staggers, layout animations, micro-interactions, scroll effects, and page transitions. Use when building or improving UI animations, adding polish, or making interfaces feel premium.
Web animations with Motion (formerly Framer Motion) for React -- gestures, scroll effects, spring physics, layout animations, SVG, micro-interactions, loading states. Use when: drag-and-drop, scroll animations, modals, carousels, parallax, page transitions, hover effects, staggered lists, loading spinners, number counters. Troubleshoot: AnimatePresence exit, list performance, Tailwind conflicts, Next.js "use client", layout in scaled containers.
Execute the next task of a free local or server-planned Altitude learning project — the learner writes every line of what is being taught, drills each new concept in a scratch file first, predicts before runs that exercise new material, and is reviewed only on what they have already learned. Use when the user says "next lesson", "let's continue the project", "next task", or invokes /next-lesson.
Use when planning a feature that has decisions to settle before building. Settles every open decision by asking the user, locks the answers into one markdown plan, in a plan file or GitHub issue, then cuts the work into as few phases as it needs, each one the user can verify.
Gate an implementation plan BEFORE agents execute it — catch phantom paths and false reuse, not typos: read the full plan markdown and its companion JSON, reality-check every file path and reuse claim against the actual repo, challenge scope, architecture, the dependency graph, and risk and test coverage, then score each section and render an APPROVE / REVISE / REJECT verdict — runs as a forked review workflow with its own reasoning budget, isolated from the main execution flow. The verdict follows only from evidence-backed findings: never rubber-stamps a non-trivial plan and never invents findings, but a genuinely clean section is allowed to pass; read-only, never edits the plan. Use when the user asks to review or gate a plan before execution.
Discovers user intent and generates a structured, step-by-step plan for model customization workflows. This skill must always be activated alongside any other skill when the user's request relates to model customization — including fine-tuning, training, building, customizing, reviewing data, or getting advice on approach, regardless of domain. Do not skip this skill even if the immediate ask is narrow (e.g., reviewing data format or a single workflow step), because planning discovers the full scope of work needed. Also activate when the user wants to resume, continue, or modify an existing plan.
Every Dub feature, plus offline search, agent-native output, and a local SQLite store no other Dub tool has. Trigger phrases: `shorten a link with Dub`, `audit my Dub links`, `find dormant Dub links`, `review Dub bounty submissions`, `Dub partner leaderboard`, `use dub-pp-cli`, `run dub-pp-cli`
Presentation creation, editing, and analysis. When Claude needs to work with presentations (.pptx files) for: (1) Creating new presentations, (2) Modifying or editing content, (3) Working with layouts, (4) Adding comments or speaker notes, or any other presentation tasks
Expert product launch strategist for SaaS and technology companies. Use when planning product launches, coordinating cross-functional launch teams, managing beta programs, creating launch communication plans, planning launch day execution, setting up post-launch monitoring, running launch retrospectives, or defining launch metrics. Covers launch tiering, internal enablement, rollback planning, and contingency strategies.
Universal prompt engineering techniques for any LLM. Use when crafting, optimizing, or reviewing prompts for AI models. Triggers on requests like “improve this prompt”, “write a system prompt”, “optimize my instructions”, “help me prompt engineer”, “audit this prompt”, “review my prompt”, or when building agentic systems that need structured prompts.
Rigorous engineering orchestrator ported from Lauren Tan's pstack (poteto-mode): reads your task, picks one of 23 playbooks (bug fix, feature, refactoring, perf, investigation, prototype, babysit, shipping, autonomous run, orchestrate, and more), routes to bundled procedures (how, why, architect, arena, swarm, interrogate, unslop, technical-writing, show-me-your-work, tdd, and others), and applies 24 engineering principles. Self-contained: no plugin install, no sibling skills required, works with any agent that reads skills.sh-format SKILL.md files. Use whenever a task needs rigor: nontrivial code changes, architecture decisions, debugging, reviews, PRs, long autonomous runs, or "work like poteto", "poteto-mode", "pstack".
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.