Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Initialize an AscendC operator project and create a compilable operator skeleton. Use when: (1) the user asks to create a new operator; (2) keywords include ascendc operator, create operator, operator directory, or operator initialization; (3) they need to quickly implement from an ascend-kernel template. This skill does more than create directories: it also outputs standard files and a checklist ready for continued development.
Use when a CAAnimation completion handler doesn't fire, spring physics look wrong on device, animation duration mismatches actual time, gesture + animation interaction causes jank, or timing differs between simulator and real hardware — systematic CAAnimation diagnosis with CATransaction patterns, frame rate awareness, and device-specific behavior
Comprehensive skill for working with Azure DevOps REST API across all services, including Boards (work items, queries, backlogs), Repos (Git, pull requests, commits), Pipelines (builds, releases, deployments), Test Plans, Artifacts, organizations, projects, security, extensions, and more. Use when implementing Azure DevOps integrations, automating DevOps workflows, or building applications that interact with Azure DevOps services
Jetpack Compose and Compose Multiplatform (KMP/CMP) architecture skill. Only use when the user explicitly mentions "compose-skill", "@compose-skill", or "use compose skill" in their message. Do NOT auto-activate based on keyword matching — this skill should only be triggered by direct user request.
Use when implementing signals in C#—[Signal] delegates, EmitSignal patterns, async signal awaiting, and event-driven architecture. Triggers on declaring, emitting, connecting, or awaiting a signal from a C# script, and on a signal crossing between C# and GDScript in either direction
Builds and deploys compiled Airflow Java SDK bundles so workers can run them. Use when the user wants to package a JVM task bundle into a JAR, asks about the `org.apache.airflow.sdk` Gradle plugin, `./gradlew bundle`, the Maven shade/BOM setup, fat vs thin JARs, the logging integration artifacts (JPL, SLF4J, Log4j 2, JUL), preview/snapshot builds, or getting the JAR onto an Airflow worker (Docker, Kubernetes, or Astro). For the task code see authoring-java-sdk-tasks; for the Airflow coordinator settings see configuring-airflow-language-sdks.
For creating standalone mind maps with content and reading/writing their themes, use dws mind; use wiki for blank mind maps and minutes for mind maps built into meeting notes. Create, draw, save, and edit standalone DingTalk whiteboards (whiteboard files). For embedded whiteboards in documents, read and write graphic content only when the containing document's nodeId and partId are already known (dws whiteboard). These two target types are independent; do not infer or switch between them. Embedded whiteboard card containers are not part of this skill. This long-tail product collection skill covers less frequently used DingTalk products: OA approval queries and processing, legal smart contracts, attendance, live streaming, DING urgent messages, Open Platform apps and MCP development, Agoal goal management, logs and daily/weekly reports, spreadsheets, Open Platform document search and OpenAPI escape hatch, DWS skill marketplace installation, organizational brain Hrbrain, native Markdown/HTML, PAT behavioral authorization, multi-organization profiles, and enterprise fitness, including DingTalk step leaderboards and sports competitions. Use when the user mentions any of these products, especially Agoal, goal management, strategy decoding, business contracts and field configuration, scorecards, OKR, goal rules and cycles, personal goals, goal templates, weekly/monthly report rule submission statistics, on-time/late/not submitted status and follow-ups; querying or creating contract registers, bulk contract import, drafting contracts from meeting notes, archiving contracts, managing contract projects/counterparties/payables; querying pending approvals, approving/rejecting/transferring/withdrawing approvals; clocking in, shift scheduling, filling daily/weekly report content; creating standalone whiteboards with content, reading/writing standalone whiteboard nodes, reading/writing embedded whiteboard graphics when nodeId and partId are known; cell read/write; searching for or installing skills; developer-console or enterprise internal apps; checking app status or deletability; app membership and role summaries; app version history; enterprise app bot settings, message modes or online status; local bot connectors; Stream connections; MCP services or tools; publishing/debugging MCP; searching the MCP marketplace; obtaining or calling published MCPs; turning OpenAPI, Swagger, Postman, curl, or HTTP APIs into MCP or making them available to an Agent; unwrapped OpenAPI; llms.txt; dws api; talent pools; employee profiles, career histories, performance; native .md files, Markdown version comparison, local draft diff, Markdown comments, native .html files, creating HTML pages, PAT authorization, switching organizations, cross-organization profiles; or enterprise step leaderboard activation, step summaries, employee/department rankings, multi-day step details, creating combined step-count/running/sports challenges, competition rules and periods, launch announcements in groups, and competition registration, participation, and achievement trends. Weekly/monthly report content entry belongs to report. Rule-level on-time, late, and missing submission statistics and follow-ups belong to Agoal. Daily/weekly report or briefing submissions with approvers, CC recipients, or approval flows belong to OA approvals. Real-time monitoring of future approval task or instance changes is not part of this skill; use dingtalk-event. After a match, use this skill's “Product Index” to identify the specific subproduct and command prefix, then follow that subproduct's instructions.
Platform-specific Docker considerations for Windows, Linux, and macOS. PROACTIVELY activate for: (1) Docker Desktop on Windows (WSL2 vs Hyper-V backends), (2) Docker Desktop on macOS (Apple Silicon, Rosetta, virtiofs), (3) native Docker Engine on Linux, (4) rootless Docker setup, (5) cross-platform image building (--platform, buildx, multi-arch manifests), (6) ARM64 vs x86_64 image selection, (7) volume performance differences (bind mount vs named volume across platforms), (8) Docker Desktop resource tuning per OS. Provides: per-platform setup steps, multi-arch build recipes, rootless setup, performance-tuning checklist, and known platform-specific gotchas.
Find and download virtually any digital resource from the internet — ebooks, academic papers, movies, TV shows, music, software, images, fonts, courses, and more. Covers both English and Chinese internet ecosystems. Includes CLI tool workflows (yt-dlp, aria2, gallery-dl, spotdl), resource site directories, cloud drive search engines (百度/阿里/夸克网盘搜索), and search techniques (Google dorks). Use when the user wants to: (1) download a video, audio, or media from a URL, (2) find and download an ebook or academic paper, (3) find and download software, (4) search for any digital resource, (5) batch download images or media from a gallery/site, (6) download torrents or magnet links, (7) find free stock assets (images, video, audio, fonts), (8) search Chinese cloud drives for resources, or (9) any task involving finding or downloading digital content from the internet.
Detect installed AI coding agents and append honest-feedback rules to their instruction files. Scans for config files (CLAUDE.md, copilot-instructions.md, .cursorrules, etc.), adapts output format per agent, and appends directives that disable sycophancy and enable constructive pushback. Use when setting up honest feedback, disabling people-pleasing, or enabling objective criticism. Triggers on honest agent, objective feedback, no sycophancy, honest criticism, contradict me, challenge assumptions, honest mode, brutal honesty
Configure passwordless SSH login to Huawei Cloud ECS instances using COC (Cloud Operations Center). Automates IAM agency authorization, SSH key pair generation, COC script deployment to one or many target ECS instances (batch), SSH connection testing, and automatic security cleanup after 60 seconds (removing keys from both local and remote). 触发词: 免密登录, COC SSH, ECS key login, SSH key deployment, passwordless SSH
Plan and execute product isolation migrations to a facade plus contract layer in PostHog, following the Visual review architecture. Use when a product still exposes internals (models/logic/views) across boundaries and needs migration toward contracts.py + facade/api.py + presentation separation, with a PR strategy that minimizes review latency and conflicts with parallel work
Write, refactor, and review Rails code using layered architecture principles from “Layered Design for Ruby on Rails Applications”. Use when writing or refactoring Rails code—models, controllers, services, jobs, mailers, policies, forms, query objects, presenters, view components, state machines, serializers, or AI/LLM features—to apply correct patterns and avoid layer violations; and when reviewing Rails code, PRs, or diffs for layer violations, fat controllers/models, anemic models, callback misuse, god objects, or specification-test failures. Triggers on “layered design”, “architecture layers”, “abstraction layer”, “specification test”, “layer violation”, “fat controller/model”, “god object”, “anemic model”, “extract service/callback/policy/concern”, “service object”, “form/policy/query/value object”, “presenter”, “view component”, “state machine”, “Active Delivery”, “callback scoring”, “Rails refactor/review”, “Rails patterns/best practices”, “archspec”, or “architecture linter/enforcement”.
Design engineering principles for making interfaces feel polished. Use when building UI components, reviewing frontend code, implementing animations, hover states, shadows, borders, typography, micro-interactions, enter/exit animations, or any visual detail work. Triggers on UI polish, design details, "make it feel better", "feels off", stagger animations, border radius, optical alignment, font smoothing, tabular numbers, image outlines, box shadows.
UX/UI audit skills, each a structured methodology: ux-audit-rethink (IxDF UX factors), nielsen-heuristics-audit, wcag-accessibility-audit (POUR framework), don-norman-principles-audit, cognitive-walkthrough, ui-design-review, and information-architecture-audit. The package also holds AI-governance and security-audit skills that this catalogue does not install.
Emulated Microsoft Entra ID (Azure AD) OAuth 2.0 / OpenID Connect for local development and testing. Use to test Microsoft sign-in locally, emulate Entra ID OIDC discovery, handle Microsoft token exchange, configure Azure AD OAuth clients, work with Microsoft Graph /me, or test PKCE/client credentials flows without calling real Microsoft APIs. Triggers include “Microsoft OAuth”, “Entra ID”, “Azure AD”, “emulate Microsoft”, “mock Microsoft login”, “test Microsoft sign-in”, “Microsoft OIDC”, “local Microsoft auth”, or any task requiring a local Microsoft OAuth/OIDC provider
Dense description covering ARP spoofing, LLMNR/NBT-NS/mDNS poisoning, DNS poisoning, MITM attacks, VLAN hopping, DHCP attacks, 802.1X/NAC bypass, IPv6 attacks. Tools: Bettercap, Responder, mitm6, Ettercap, Wireshark. MITRE T1557, T1040. Use when conducting internal network assessments or testing Layer 2/3 attack surface.
Design production-grade screens in Penpot from a brief, as a senior visual designer — reusing the existing design system (tokens + components) and assembling section by section, never one-shot. Use to create a screen/page/landing/dashboard from a description. NOT for translating existing code (use penpot-build-from-code). Triggers: 'design a dashboard', 'create a landing page', 'design this app screen', 'build a UI from this brief', 'design a settings page', 'mock up a screen in Penpot'.
Build positioning and a narrative where the developer is the hero and a real trend is the villain. Use when messaging describes the product instead of the problem, sounds like every competitor, or lacks urgency because nothing is at stake
Write and iteratively refine PRD/需求文档 with a story-driven structure and strict staged confirmations (journey map alignment, per-story single-point confirmation, final generation gate). Use when the user asks to 梳理/撰写/完善 PRD、需求文档、用户故事、验收标准，并希望用 ASCII 线框图与 Mermaid（流程图/状态图/时序图）来减少歧义、共同完成文档。
Adversarial analysis deploying parallel expert agents to stress-test ideas, strategies, and plans — attacks atomic claims, then steelmans and counter-argues, with severity-ranked findings and remediation; also tests an idea or investment against 11 persistent world models from 6 months to 50 years (Fast/Standard/Deep), and refreshes and reports on those models. USE WHEN red team, attack idea, counterarguments, critique, stress test, devil's advocate, find weaknesses, break this, poke holes, strongest objection, world threat model, world model, test idea against the future, how will this hold up in 10 or 20 years, time horizon, long-term risk, update world models, show world models. NOT FOR collaborative debate to find best path (use Council), or threat modeling your own estate (ThreatModel)
Surface past learnings relevant to the current task before starting work. Searches correction history, recalls past mistakes, and applies prior patterns. Use when starting a task, saying "what do I know about", "previous mistakes", "lessons learned", or "remind me about".
Draft professional, empathetic customer-facing responses adapted to the situation, urgency, and channel. Use when responding to customer tickets, escalations, outage notifications, bug reports, feature requests, or any customer-facing communication.
Use for Roblox persistent data and cross-server state design: choosing between DataStoreService, OrderedDataStore, MemoryStoreService, and MessagingService; designing save and load flows, schema shape, versioning, metadata, retries, quotas, observability, and concurrency-safe coordination across servers.
Generate academic presentation-style HTML slide decks and browser reports from PDFs, structured text, Markdown, paper summaries, outlines, or research notes. Use whenever the user wants to convert a scientific paper PDF directly into an interactive HTML report, clickable web presentation, shareable browser deck, or PPT-like HTML output with figures, structured insights, section pages, and offline deck-directory output. In the Aut_Sci_Write suite, choose this skill when the user wants HTML/web output; choose sci-ppt when they explicitly need a .pptx file.
Fixes security and trust issues. Configures HTTPS, HSTS, Content-Security-Policy, X-Content-Type-Options, frame protection, CORS and Referrer-Policy, secures WebMCP tool definitions and auto-submit forms, and publishes OAuth authorization server and protected resource metadata so AI agents and platforms can trust and authenticate with a website. Use when asked to "fix security headers", "add HSTS", "configure CSP", "fix HTTPS", "fix CORS", "add Referrer-Policy", "add OAuth discovery", "add oauth-protected-resource", "make auth agent-ready", "improve security score", or any security header or agent auth task.
This skill should be used when the harness, scaffold, workflow, or optimizer itself is the optimization target: recursive self-improvement (RSI) loops, meta-harnesses, self-improving harnesses that mine their own failures and propose bounded edits, evolutionary or population-based search over agent scaffolds, acceptance gates for self-modifying systems, and agentic context evolution where the mechanism that produces context is versioned and evolved. Route governance of a single autonomous loop (locked surfaces, durable logs, rollback, novelty gates, approval boundaries) to harness-engineering, measurement and quality-gate design to evaluation, judge design to advanced-evaluation, and remote sandbox infrastructure to hosted-agents.
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.