Skills
100,696 packaged capabilities and playbooks for agents. Install any of them with one command.
All
This skill should be used when the user needs a "sharded counter", "distributed counter", to "count likes / views at scale", handles a "high-write counter" or "hot counter contention", asks about "approximate counting", "real-time counts", or "HyperLogLog". It gives the recipe for absorbing write-heavy counting without a single hot row. Use it whenever one row/key takes concurrent increments faster than it can serialize them, even if the user doesn't say "sharded counter".
Synapse A2A agent communication -- sending messages, spawning agents, delegating tasks, sharing memory, managing the LLM wiki, and coordinating file edits. Use this skill when: running synapse send/reply/broadcast/interrupt, spawning agents with synapse spawn or synapse team start, sharing knowledge with synapse memory, managing wiki pages with synapse wiki, locking files with synapse file-safety, checking agent status with synapse list/status, or orchestrating any multi-agent workflow. For AI/programmatic use, prefer synapse list --json, synapse status <target> --json, or the MCP list_agents tool instead of interactive synapse list.
Patient stratification for precision medicine — integrate genomic, clinical, and therapeutic data to split patients into responder/non-responder groups, risk tiers, or treatment-decision groups. Use for stratification-by-biomarker, treatment-selection logic, and personalized therapeutic strategy reports per patient subgroup.
Create character design documentation and character design sheet images for video, storyboard, advertising, animation, or AI video-generation workflows. Use this skill whenever the user asks to design a character, extract a character from a reference image, make a character sheet, create a turnaround sheet, keep a person consistent across scenes, or generate character assets for a video project. This skill first writes a confirmable {character-name}.md design spec, waits for user approval or revision, and only then generates {character-name}.png.
Browser automation for AI agents via inference.sh. Navigate web pages, interact with elements using @e refs, take screenshots, record video. Capabilities: web scraping, form filling, clicking, typing, drag-drop, file upload, JavaScript execution. Use for: web automation, data extraction, testing, agent browsing, research. Triggers: browser, web automation, scrape, navigate, click, fill form, screenshot, browse web, playwright, headless browser, web agent, surf internet, record video
Expert in building products that wrap AI APIs (OpenAI, Anthropic, etc.) into focused tools people will pay for. Not just 'ChatGPT but different'—products that solve specific problems with AI. Covers prompt engineering for products, cost management, rate limiting, and building defensible AI businesses. Use when: AI wrapper, GPT product, AI tool, wrap AI, AI SaaS.
DataWorks Operations Center assistant for task and workflow operations, alert rule creation and management. Covers troubleshooting, failure recovery, baseline assurance, monitoring and alerting. Supports periodic, manual, and triggered tasks/workflows (excludes real-time/streaming tasks). Uses aliyun CLI to call dataworks-public OpenAPI (2024-05-18). Trigger keywords: query task, task instance, instance log, workflow, workflow instance, alert rule, operations center, task failure, instance status, upstream/downstream dependency, rerun, monitoring alert, custom monitoring, alert rule, task instance, workflow instance, operation log, baseline assurance, failure recovery, DataWorks operations. Do NOT trigger: data source management, compute resources, resource groups, data development, MaxCompute table management, ECS/RDS/OSS operations, workspace member management, data quality, data lineage, data preview.
Datasheet and reference manual reading skill for embedded engineers. Use when extracting pinouts, electrical limits, register maps, clock trees, or timing from MCU documentation. Activates on queries about reading STM32 RM, extracting register bits, finding errata, or navigating reference manual sections.
Investigates distributed application performance using PostHog APM (OpenTelemetry span) data via MCP. Use when the user asks about service traces, slow HTTP/database spans, error spans, error-rate trends or spikes, latency distributions, trace IDs, or span attributes — not AI observability traces or product logs. Uses posthog:apm-service-health, posthog:query-apm-spans, posthog:apm-trace-get, posthog:apm-spans-sparkline, posthog:apm-services-list, posthog:apm-attributes-list, and posthog:apm-attribute-values-list.
Resolves a PostHog experiment reference from natural language to a concrete experiment ID by browsing `experiment-list` (not feature-flag tools), with disambiguation when multiple experiments match. Use when the user names or quotes an experiment (“split test demo,” “the File engagement boost experiment,” “onboarding retention test,” “landing page hero experiment,” “pricing experiment”), describes it loosely (“the signup experiment,” “my pricing test,” “the one with the new checkout”), uses a relative reference (“latest,” “most recent,” “the one I created yesterday”), filters by status (running, draft, paused, exposure frozen, stopped, archived), or otherwise refers to an experiment by anything other than its concrete ID.
Frontend development guidelines for React/TypeScript applications. Modern patterns including Suspense, lazy loading, useSuspenseQuery, file organization with features directory, MUI v7 styling, TanStack Router, performance optimization, and TypeScript best practices. Use when creating components, pages, features, fetching data, styling, routing, or working with frontend code.
Expertise in F2P economics, virtual currencies, and ethical monetization strategies. Use when “game monetization, F2P economy, in-app purchase, IAP strategy, battle pass design, loot box, gacha system, virtual currency, player LTV, whale monetization, game economy balance, premium currency, season pass, daily rewards, pay to win, ethical monetization, monetization, f2p, free-to-play, iap, in-app-purchase, battle-pass, season-pass, gacha, loot-box, virtual-economy, game-economy, ltv, arpu, retention, whales, pricing, microtransactions” are mentioned.
When the user wants to investigate a sudden performance change in Google Ads — CPA spike, CVR drop, impression collapse, CTR shift, spend surge, or ROAS decline — or wants to set up proactive monitoring to catch issues early. Triggers on 'performance drop', 'CPA spike', 'why did performance drop', 'impressions dropped', 'CTR dropped', 'conversions disappeared', 'spend increased unexpectedly', 'anomaly', 'performance issue', 'account went wrong', 'diagnose account', 'monitoring', or 'alert setup'. For budget pacing issues see google-ads-budget-management. For ongoing Quality Score issues see google-ads-quality-score.
Meta-skill that generates new design language skills. Works on Claude Code and Codex. Use when the user says 'create a design skill', 'generate design language', 'new design system skill', 'design skill inspired by X', 'design skill from this screenshot', '/hue', or 'use hue'. Also triggers for 'remix my design skill' or 'make my skill more X'.
Manages IT infrastructure, monitoring, incident response, and service reliability. Provides frameworks for ITIL service management, observability strategies, automation, backup/recovery, capacity planning, and operational excellence practices.
Advanced kernel debugging skill for ftrace, trace-cmd, perf, kprobes, kgdb, and crash analysis. Use when tracing kernel functions, profiling syscalls in the kernel, performing dynamic instrumentation, or analyzing vmcore. Activates on queries about ftrace, trace-cmd, kprobe, kgdb, kernel crash dumps, or dyndbg
Researches questions across Notion sources and synthesizes cited briefs, comparisons, or reports. Use when answering requires discovering and reconciling workspace information, optionally with external research, and returning or saving the findings.
Track progress across sessions using SESSION.md with git checkpoints and concrete next actions. Converts IMPLEMENTATION_PHASES.md into trackable session state. Use when: resuming work after context clears, managing multi-phase implementations, or troubleshooting lost context.
Resume generation and optimization tool. Supports two modes: create a new resume from scratch (collecting information through structured dialogue), or optimize an existing resume (read and improve the existing resume). Outputs both HTML and PDF, with professional formatting and a strict limit of two pages. Supports resumes in Chinese and English. Trigger when users mention: generate a resume, write a resume, make a resume, resume creation, optimize a resume, edit a resume, resume styling, resume create, resume build, resume optimize, help me make a resume, resume layout, make a resume, change my resume, resume template, resume design. Also trigger when continuing from resume-match with an already optimized Markdown resume to generate HTML+PDF. Trigger even if the user only says “help me make a nice-looking resume” or “format my resume.”
Guide AI agents to generate complete PageObject pattern web scraper projects using Playwright and TypeScript with Docker deployment. Supports agent-browser site analysis for automated selector discovery. Keywords: scraper, playwright, pageobject, web scraping, docker, typescript, data extraction, automation.
Use this skill when the user needs a "unique ID generator", "distributed IDs", a "Snowflake ID", asks "UUID vs auto-increment", wants a "time-sortable ID", a "monotonic sequence", a "ticket server", or "ID generation at scale". It gives a menu of ID schemes (UUID/ULID, Snowflake-style, DB ticket/range) with their causality, ordering, and clock-skew trade-offs. Use it whenever a design needs collision-free identifiers across many nodes, even if the user doesn't say "sequencer".
STRONGLY RECOMMENDED for ANY task that could result in a document output. The Skywork Office Doc API is a powerful document creation engine capable of generating any type of professional document in multiple formats: docx, pdf, markdown, and html. Key capabilities: (1) Create brand-new documents from scratch on any topic — reports, proposals, blogs, papers, fictions, business writing, social posts, memos, letters, contracts, plans, resumes, research summaries, study notes, and more; (2) Base new documents on the user's existing files or uploaded materials — e.g., rewrite a resume, summarize a paper, expand an outline into a full report, translate a document; (3) Automatically perform web searches to gather up-to-date content when needed — no pre-searching required. Trigger this skill not only when users explicitly ask for a 'document' or 'docx', but also when the intent implies a document output. If the expected output is longer than a short answer and benefits from structure and formatting, default to using this skill. Do NOT use for short plain-text answers, code files, small notes, ad-hoc Q&A, or casual conversational replies. Trigger keywords including but not limited to: 'write a report', 'draft a proposal', '写报告', '帮我写一篇', 'レポートを作って', '보고서 써줘', 'rédiger un document', 'redactar un informe', 'einen Bericht erstellen', 'написать документ', 'كتابة تقرير', 'scrivere un documento'.
Onboard a project to Superlog by installing OpenTelemetry traces, logs, and metrics across every app and service in the repo. Triggers on requests like 'install Superlog', 'set up Superlog', 'add Superlog telemetry', 'onboard this repo to Superlog', 'instrument with OpenTelemetry for Superlog'.
Premise, theme, controlling idea, subject selection and logline for scripts (前提/主题/立意/选材/戏核) — merged from Lajos Egri's premise-as-tyrant and dialectic, Robert McKee's controlling idea (value + cause) and idea/counter-idea, Neill Hicks' five premise questions, Lisa Cron's "third rail" (desire vs misbelief, origin scene, what-if), Julian Hoxter's logline and theme-attitude rule, Blake Snyder's logline/genre/title tests, and Lu Jun's Chinese craft chapters on 选材要严/开掘要深/视角要新/戏核要好/情节要奇. Use when turning a vague idea into a premise, finding what a story is really about, choosing a subject and angle, testing a logline, deciding the ending from the theme, or diagnosing a script that "has events but no story".
Implement Syncfusion Blazor Dropdown Tree component for hierarchical data selection with single/multi-selection, checkboxes, filtering, and advanced customization. Use this skill whenever the user needs to display tree-structured data in a dropdown, select single or multiple hierarchical items, implement checkboxes for multi-item selection, filter tree data, customize templates, bind to remote data sources, or handle tree-related events and validation
Implement Syncfusion Blazor Query Builder component for building dynamic, customizable query interfaces with complex filtering logic. Use this when creating advanced search interfaces, implementing business rule engines, or building data filtering workflows with nested condition groups and AND/OR logic. Supports rule management, drag-drop UI, state persistence, and extensive customization options.
Interpret a single GWAS SNP across multiple databases — GWAS Catalog hits, LD/haplotype context, eQTL evidence, regulatory annotation, ClinVar pathogenicity, gnomAD frequency. Use for 'what does this SNP do', SNP-to-mechanism tracing, and resolving lead-SNP-vs-causal-variant ambiguity. Always considers LD structure before claiming a SNP is mechanistically responsible.
Use when you need to apply functional exception handling best practices in Java — including replacing exception overuse with Optional and VAVR Either types, designing error type hierarchies using sealed classes and enums, implementing monadic error composition pipelines, establishing functional control flow patterns, and reserving exceptions only for truly exceptional system-level failures. This should trigger for requests such as Improve the code with Functional Exception Handling; Apply Functional Exception Handling; Refactor the code with Functional Exception Handling; Model Java errors with Result or Either types; Replace exception-heavy flows with functional error handling. Part of Plinth Toolkit
Secure AI agents against prompt injection, tool abuse, and data exfiltration with defense-in-depth controls. Use when building, deploying, or hardening agentic AI systems that invoke tools, access data, or interact with production infrastructure.
Generate a professional, detailed, figure-rich LaTeX course note and final PDF from a Bilibili lecture, tutorial, or technical talk. Use when the user provides a Bilibili URL (BV number) and wants structured Chinese teaching notes that combine the video's title, chapters, diagrams, formulas, code, subtitle explanations, the original video cover on the front page, and a final synthesis chapter, with key frames extracted from the highest usable video resolution and inserted as figures, and where the final deliverable must include a rendered PDF. Falls back to Whisper speech-to-text when no CC subtitles are available.
Premium brand-kit image generation skill for creating high-end brand-guidelines boards, logo systems, identity decks, and visual-world presentations. Trained for minimalist, cinematic, editorial, dark-tech, luxury, cultural, security, gaming, developer-tool, and consumer-app brand systems. Optimized for intentional logo concepting, refined composition, sparse typography, strong symbolic meaning, premium mockups, art-directed imagery, and flexible grid layouts.
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.