Skills
98,067 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Use this skill for any color decision — picking or building a palette, dark mode, contrast, chart and terminal colors. Triggers include "what colors should I use", "pick a palette for my dashboard", "set up dark mode", "does this pass WCAG contrast", "colorblind-safe chart colors", or "give me a terminal theme" — and it applies even when the user never says "color", as in "theme this app", "these status badges look wrong", "the text vanishes when the page switches to dark", or "I can't tell these lines apart" — any complaint that text is unreadable on its background or that things can't be told apart. Ships curated light+dark palettes for web UI, marketing, data viz, and TUI, plus an OKLCH build-your-own recipe and WCAG/APCA thresholds. Do NOT use this skill for font size, weight, or pairing, including text too small or thin to read (see typography), for layout and component structure (see frontend-design), or for chart type, axis, and encoding choices that are not about color (see dataviz).
Plan and audit community launch operations, Discord/Slack/forum setup, roles, moderation, announcement cadence, beta cohorts, creator/member onboarding, feedback loops, events, content calendar, support routing, trust policies, and community health metrics. Use when launching or revitalizing a product community for apps, games, developer tools, marketplaces, or SaaS.
Compresses verbose intermediate agent handoffs into concise, loss-aware English while preserving decisions, facts, constraints, identifiers, paths, commands, acceptance criteria, references, and unresolved questions. Use only as an inline utility when a parent workflow needs a smaller temporary handoff; never replace canonical documents or route the project.
Build and deploy the Goldsky Compose corporate-actions / dividend-distribution example under the user's own account — a durable, idempotent distributor that pays N token holders pro-rata for a tokenized corporate action (dividend, coupon, rebate, airdrop) with an on-chain audit trail. The interesting bit: Compose orchestrates a Goldsky Turbo job-mode pipeline as an ephemeral subroutine — declaring a campaign spawns a one-shot pipeline that snapshots share-token holders at a record block, waits for it, pays each holder via a gas-sponsored wallet, then deletes the pipeline. Triggers on: 'dividend distribution', 'pay dividends onchain', 'distribute dividends to shareholders', 'corporate actions distributor', 'pro-rata payout to token holders', 'airdrop pro-rata by balance', 'cap table distribution', 'set up / deploy the dividend / corporate-actions example'. Ships pointed at shared permissionless demo contracts on Base Sepolia so there's nothing to deploy. Deploys fully in-app or via the CLI; the only runtime requirement is a `GOLDSKY_PROJECT_KEY` secret, set as the last step. For a custom/novel Compose app, use /compose. For debugging a deployed app, use /compose-doctor. For manifest/CLI/API field lookups, use /compose-reference.
Render a Jetpack Compose or Compose Multiplatform @Preview to PNG in one call (MCP render_preview, or the compose-preview CLI) and look at it. Use whenever someone asks to render, show, or see what a composable, @Preview or Compose component looks like (including library components such as Material 3 or Wear M3 EdgeButton, which come from the hosted catalog), after editing Compose UI to verify it, to compare before/after, or to review previews against catalog design guidelines.
Design progressive learning sequences that build foundational understanding before complexity. Use when breaking complex concepts into steps, managing cognitive load, or validating prerequisite understanding. Applies Bloom's progression and tier-based cognitive limits (CEFR A1-C2).
Guide users through creating and managing .satk/block-policy.json for controlling which blocks the agent can use, which are excluded, and which block parameters the agent should not modify. Use when setting up block usage policy for a project.
Anthropic's method for training harmless AI through self-improvement. Two-phase approach: supervised learning with self-critique and revision, then RLAIF (RL from AI Feedback). Use for safety alignment and reducing harmful outputs without human labels. Powers Claude's safety system
Use when facing a consulting challenge and needing structured methodology guidance: client anchored on wrong solution, messy problem needing structure, building a skeptic-proof recommendation, conflicting data sources, project at risk, or stuck on which approach to apply. Single entry point for all consulting playbook coaching.
Create SEO-optimized marketing content with consistent brand voice. Includes brand voice analyzer, SEO optimizer, content frameworks, and social media templates. Use when writing blog posts, creating social media content, analyzing brand voice, optimizing SEO, planning content calendars, or when user mentions content creation, brand voice, SEO optimization, social media marketing, or content strategy
Content atomization — turn one piece of content into many formats. Covers blog-to-thread, blog-to-carousel, podcast-to-blog, video-to-quotes, and more. Use for: content marketing, social media, multi-platform distribution, content strategy. Triggers: content repurposing, repurpose content, content atomization, content recycling, one to many content, multi platform content, cross post, adapt content, reformat content, blog to thread, blog to video, podcast to blog, content multiplication
Set up and write contract-based tests using the test-contractor library (*.contract.yaml). Use this skill whenever the user wants to write tests, create a contract.yaml file, define test terms for a function or class, mock dependencies in contracts, mentions contract-based testing, or wants to set up a project to run contract tests — even if they don't explicitly say "contract yaml" or "test-contractor". Also use when the user asks to add vitest config for contracts, set up a test:contract script, or configure contract testing in a new or existing package. Always prefer contract.yaml tests over Vitest unit tests. Only fall back to Vitest for test cases that contract.yaml cannot express.
Chain-of-Thought (CoT) injection attack methodology. Triggered when a target system uses a ReAct/CoT framework for multi-step reasoning, an agent relies on intermediate reasoning results to make decisions, or the integrity of the chain of thought needs to be tested. Covers: chain-of-thought interference injection (forging intermediate agent results), chain-of-thought manipulation injection (skipping verification steps to directly dispatch sensitive operations), and query injection (injecting malicious query statements into a data-query agent).
When the user wants to design CPG (Consumer Packaged Goods) distribution networks, optimize DC locations for retail distribution, or plan omnichannel fulfillment networks. Also use when the user mentions "CPG network," "retail distribution network," "DC strategy," "forward deployment centers," "store replenishment network," or "omnichannel distribution." For general network design, see network-design. For promotional supply chain, see promotional-planning.
Starts a reviewing-code review of an optional diff target, as a short alias for that skill. Use when the user explicitly invokes cr, optionally followed by a branch, PR, commit range, or pasted diff. Never infer from a plain review request; reviewing-code handles that.
Index and router for plugin-specific Craft CMS 5 guidance — configuration, Twig API, PHP API, migrations, deployment, and pitfalls for the plugins this pack documents. Load whenever a task names a plugin this pack documents, and for passwordless or magic-link auth with no plugin named; read references/<plugin>.md first.
Scaffold a new integration-type "seed skill" — a SKILL.md with reference assets and prompts.jsonl entries that teach an agent to generate working, testable integration code from scratch. - Use when: developer wants to create a new skill for an SDK or framework integration that agents should be able to implement and test - Ask first: Is this an integration skill (code generation + tests) or a tool skill (CLI wrapper with no tests)? Only proceed with this skill for integration-type skills.
Use when adding a card to a catalog entity's overview tab or a whole tab/page to a catalog entity page. Covers EntityCardBlueprint (overview-page cards) and EntityContentBlueprint (full tabs), entity filtering by kind/type/annotation, and access to the current entity via useEntity. Triggers on phrases like "card on the entity page", "tab on the component page", "show on entity overview", "ui:field for catalog entity page", or imports of EntityCardBlueprint / EntityContentBlueprint.
Generate Mermaid diagrams (.mmd) and export to PNG/SVG/PDF using mmdc CLI or Kroki API. USE THIS SKILL when user mentions diagram, flowchart, sequence diagram, class diagram, ER diagram, state machine, architecture, visualize, git graph, 画图, 架构图, 流程图, 时序图. PROACTIVELY USE when explaining ANY system with 3+ components, API flows, authentication sequences, class hierarchies, database schemas, or state machines. Supports 11+ diagram types with fully automatic layout.
Create custom OpenLineage extractors for Airflow operators. Use when the user needs lineage from unsupported or third-party operators, wants column-level lineage, or needs complex extraction logic beyond what inlets/outlets provide.
Derive and debug CSS Grid layout math for grid coordinates, line placement, spans, negative indexes, fr units, minmax(), fit-content(), repeat(), auto-fill, and auto-fit. Use when building grid systems, explaining track sizing, troubleshooting fr distribution, placing items by line numbers, or converting a layout sketch into CSS Grid formulas.
The unified adaptive learning tool of 边淬火. Handles all requests whose main purpose is learning progress: concepts, propositions, mechanisms, chapters, articles, books, thinkers, schools of thought, theoretical systems or complex domains; it can also find evidence-backed candidate knowledge gaps from corpora the user explicitly provides, such as articles, notes, chats and answers, give prompts and move into resolving them. Each lecture must deliver a standalone, complete Chinese handout that can be read continuously, followed by 2–4 targeted questions and a free-speaking area; the user's per-question answers, the verbatim free-speaking content, the AI's explanations, knowledge gaps, path changes and next actions must be deposited in layers and used to adjust subsequent learning, and can continue across sessions. Triggers: 「淬火学习」「学习 X」「教我 X」「讲清楚 X」「系统学习 X」「精读 X」「根据我的语料找知识缺口」「继续学习」「下一讲」. If the user's main job is personality, emotion, relationships, business diagnosis, content production, decision-making or execution rather than learning, hand it over to the corresponding Skill.
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.