Skills
98,067 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Use when complex problems require systematic step-by-step reasoning with ability to revise thoughts, branch into alternative approaches, or dynamically adjust scope. Ideal for multi-stage analysis, design planning, problem decomposition, or tasks with initially unclear scope.
Use when building, editing, validating, testing, or debugging an n8n workflow through the n8n-mcp MCP server—designing a flow, configuring a node, writing an expression or Code node, wiring credentials, or fixing one that misbehaves. The entry-point skill for the n8n-mcp-skills pack: it routes you to the right specialist skill, gives working knowledge of every n8n-mcp tool from turn one, and states the rules that keep workflows from breaking in production. Always consult it first on any n8n, workflow, node, or automation task—even a quick one-off, and even when the user names no skill—because n8n's surface drifts between versions and the specialist skills prevent silent failures.
Databricks CLI operations and the parent/entry-point skill for Databricks CLI use: authentication, profile selection, and bundles. Load this first for CLI, auth, profile, and bundle tasks, then load the matching product skill. For finding or exploring data, answering questions about the data, or generating SQL, load the databricks-data-discovery skill (it routes to Genie One). Contains up-to-date guidelines for Databricks-related CLI tasks.
One book → one clear “viewfinder” concept image → one 2050 library card (PNG). The viewfinder captures the angle from which the author examines an issue and the image they see; the card includes the real cover, author portrait, and bibliographic details. The viewfinder block uses Feynman-style explanations to make the concept image clear and accurate; the illustration block uses AI image generation to depict it, with Jijang as the fixed protagonist (generated from a reference ink portrait and recognizable as him). Illustration style: picture-book feel in the style of Yoshida Seiji (everyday spaces in another world, warm slanting light, soothing yet refined). Light optical-glass card body, accent color dynamically extracted from the cover, adaptive aspect ratio. Close the book and remember this image, and the reading won’t be in vain. Use when the user says “取景框卡,” “图书馆卡,” “library card,” “书卡,” “铸书卡,” “一本书一句话一张卡,” “/ljg-library,” or provides a book title and wants it distilled into one collectible card. NOT FOR structural book analysis (use ljg-book), text-only quotes (use ljg-card -b), infographics (use ljg-card -i), or visual notes (use ljg-card -v).
Create a packaging design brief — structure, visual direction, hierarchy, materials, and unboxing experience. Use when the user says "packaging design", "packaging brief", "product packaging", "packaging direction", "label design", "packaging identity", "unboxing experience", "packaging look and feel", "brief my packaging designer", "packaging for my product", "packaging strategy", "retail packaging", "e-commerce packaging", "what should my packaging look like", or is briefing a designer, agency, or manufacturer on packaging requirements.
Comprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Claude needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4) Adding comments, or any other document tasks
Instrument an application with Sentry — detect the platform, install and initialize the SDK if needed, and wire up any signal — error monitoring, tracing/performance, logging, metrics, profiling, session replay, user feedback, cron check-ins, uptime monitors for the deployed app, and AI/LLM monitoring (agent runs, token cost, and conversations for OpenAI, Anthropic, Vercel AI, LangChain, Google GenAI, Pydantic AI, Laravel AI, Mastra, Eve, Flue, Pi Durable, the Cloudflare Agents SDK, and Workers AI). Use to add Sentry to a project or to capture more than errors.
Search Slack for interpreted organizational context -- decisions, constraints, and discussion arcs -- and produce a synthesized research digest with cross-cutting analysis. Use when the user says 'search slack for', 'what did we discuss about', 'slack context for', or 'what does the team think about'. Differs from slack:find-discussions, which returns raw message results without synthesis.
Creates integration test cases targeting Oracle for .NET data access artifacts. Tests capture Oracle expected behavior as the authoritative baseline; they are written once and later ported to PostgreSQL by migrating the test project in Phase 6. Use only during Phase 3, before any PostgreSQL migration work has begun. Do not invoke during Phase 6 or against a project that has already been migrated.
Design experiment plans with progressive stages—initial implementation, baseline tuning, creative research, and ablation studies. Plan baselines, datasets, hyperparameter sweeps, and evaluation metrics. Use when planning experiments for a research paper
Build self-serve acquisition and expansion motions. Use when deciding PLG vs sales-led, optimizing activation, driving freemium conversion, building growth equations, or recognizing when product complexity demands human touch. Includes the parallel test where sales-led won 10x on revenue.
Help users define problems clearly before jumping to solutions. Use when someone is scoping a new feature, validating a product idea, struggling to articulate what they're building, or falling into the “shiny object trap” with new technology.
Decide whether to act on a caller's (喊单人) calls. Scores one call-out wallet on two axes — SELECTION (do the coins he calls actually run) and POST-CALL CONDUCT (does he sell into the attention his own call created) — then returns a follow / don't-follow verdict. The conduct axis is proved from the caller's own per-trade ledger, not from any self-reported field: for each sampled coin it compares the first buy, the call timestamp, and every sell after it, so "he dumped on his followers" is a measured sequence rather than an inference. USE THIS SKILL WHEN the user asks whether a caller is worth following: 这个喊单的能不能跟, 这个喊单号靠不靠谱, 喊单信誉分, 喊单评分, 他喊完会不会砸盘, 他是不是喊完就卖, 跟他的单能不能吃到, caller 评分, is this caller trustworthy, can I follow this caller's calls, does he dump after calling, call-out credibility score, score this caller. Accepts either a GMGN profile handle (twitter_username, e.g. the last path segment of a gmgn.ai/profile/<handle> link) or a wallet address; a handle is resolved to one (chain, wallet) pair in Step 0 and the resolution is always echoed. The same questions in any other language route here — match on meaning, not wording. DO NOT USE THIS SKILL for raw call-out data — "这个钱包 喊过什么单", "最新喊单", "谁在喊这个币", "KOL 在喊什么", the feed, the field dump, the 30d summary card — that is gmgn-callout, which owns every read of the callout endpoints and returns them untouched; this skill calls it and adds a verdict. Nor for a caller judged as a trader rather than as a caller ("这个钱包能不能跟单", "跟单评分", copy-trade backtest) — that is gmgn-wallet-analysis / gmgn-wallet-score, which read his trades and know nothing about his calls. Nor for a token the caller mentioned ("他喊的那个币安全吗", 尽调, rug check) — that is gmgn-contract-dd. Nor for a token creator's next launch ("这个 dev 的新盘能不能买") — that is gmgn-dev-score. A bare wallet address with no call-out intent stays with gmgn-wallet-analysis; a bare token address stays with gmgn-contract-dd. This skill needs the word 喊单 / caller / call-out in the ask to own it. TIE-BREAKS, because three words are shared with neighbouring skills: 喊单 paired with 能不能跟 / 信誉 / 砸盘 / 评分 routes HERE, while 喊单 paired with 喊过什么 / 最新 / 谁在喊 / 列表 routes to gmgn-callout — the first group asks for a judgement, the second for rows. 能不能跟 without 喊单 is about a wallet as a trader and stays with gmgn-wallet-analysis / gmgn-wallet-review. 信誉 without 喊单 is a launcher question and stays with gmgn-dev-score. PRESET PHRASES this skill is written to answer, each carrying the call-out intent word: 「帮我看看 [喊单人ID或地址] 这个喊单的能 不能跟」「[喊单人ID] 喊的币能跟吗」「查一下 [钱包地址] 的喊单信誉」「这个喊单号喊完会不会 砸盘:[地址]」. WHAT IT CANNOT ANSWER, so the ask goes elsewhere: whether any one called coin is safe to buy (gmgn-contract-dd); what he is calling right now (gmgn-callout global); whether he is a good trader to copy (gmgn-wallet-score); who else called the same coin or whether he called it first (no per-token caller list exists in the CLI, so it is not computable at all); how much he staked on a call (readable only in the unfilterable global feed); anything older than 30 days (callout record covers 30d only).
When the user wants to analyze how their social media posts are performing. Also use when the user mentions 'analytics,' 'performance,' 'how did my posts do,' 'engagement,' 'impressions,' 'what's working,' 'post metrics,' 'my best posts,' or 'why isn't this post performing.' Uses BlackTwist analytics when available, works from user-provided data otherwise. For audience growth specifically, see audience-growth-tracker-sms. For pattern detection, see content-pattern-analyzer-sms. For actionable next steps, see optimization-advisor-sms.
Creates reusable, source-traced research reports and coordinates provider/mode lanes with original exports. Use for deep research, research reports, literature reviews, market or company research, and ChatGPT/Kimi/UniFuncs research routes. Technology choice uses tech-selection; competitor code uses competitors-analysis.
Runs a declarative strategy spec over OHLCV bars with the manifoldbt Rust engine, pairs the fill log into round trips, and emits the eight inputs the backtest-expert skill scores. Use when the user wants to execute a backtest, measure a rule they have described, obtain win rate / average win / average loss / max drawdown from real bars, or feed backtest-expert with measured numbers instead of estimates.
Build professional financial services data packs from various sources, including CIMs, offering memorandums, SEC filings, web search, or MCP servers. Extract, normalize, and standardize financial data into investment committee-ready Excel workbooks with consistent structure, proper formatting, and documented assumptions. Use for M&A due diligence, private equity analysis, investment committee materials, and standardizing financial reporting across portfolio companies. Do not use for simple financial calculations or working with already-completed data packs
Discovers all projects in a .NET solution, classifies each for Oracle-to-PostgreSQL migration eligibility, and produces a persistent master migration plan. Use when starting a multi-project Oracle-to-PostgreSQL migration, creating a migration inventory, or assessing which .NET projects contain Oracle dependencies
Write and review unit tests for Vue 3 + TypeScript + Vitest + Pinia codebases. Use when creating or updating tests for components, composables, and stores; mocking Pinia with createTestingPinia; applying Vue Test Utils patterns; and enforcing black-box assertions over implementation details.
Analyzes user-requested Screenpipe history windows to detect repeated research workflows, match existing scientific skills, and stage new skill drafts or composition recipes for review. Requires a reachable Screenpipe HTTP API, normally on localhost:3030. Detection and embedding inference run locally; the selected LLM receives redacted app/title cluster summaries and matched skill descriptions. Use only when the user explicitly asks to analyze their recent work and propose skills.
Analyze how competitors present their brand — identity, messaging, positioning, voice, and visual style — to find gaps and opportunities. Use when the user says "competitor brand analysis", "analyze competitor brands", "how do competitors position themselves", "competitor messaging", "competitor identity", "what are competitors saying", "benchmark competitors", "competitive brand landscape", "how do we compare to competitors", "competitive audit", or wants to understand the brand landscape before defining their own position.
How to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules across hardware and parallelism modes, container conventions, monitoring, and per-rank failure diagnosis.
Brev instance operating guidance for NeMo-RL agents working in /home/ubuntu/RL with limited workspace disk, a larger /ephemeral volume, and optional /home/ubuntu/RL/.env secrets. Use when running nemo-rl-auto-research campaigns, experiments, training jobs, model or dataset downloads, shared cache-heavy commands, log-producing runs, checkpoint generation, W&B or Hugging Face authenticated workflows, or any workflow that may create large files on Brev.
Analyze the bond futures basis by pricing futures, identifying the cheapest-to-deliver, and comparing with yield curves to assess delivery option value and basis trading opportunities. Use when analyzing bond futures, computing the basis, identifying CTD bonds, calculating implied repo rates, or evaluating basis trades.
Show how much context window context-mode saved this session. Displays token consumption, context savings ratio, and per-tool breakdown. Read-only — shows stats only, no reset capability. To wipe the knowledge base entirely, use ctx_purge instead. Trigger: /context-mode:ctx-stats
When the user wants platform-specific tactical guidance for LinkedIn, Twitter/X, Threads, or Bluesky. Also use when the user mentions 'LinkedIn strategy,' 'Twitter strategy,' 'Threads strategy,' 'Bluesky strategy,' 'algorithm,' 'what works on LinkedIn,' 'cross-posting,' 'platform differences,' 'adapt my content,' or 'which platform should I focus on.' For overall content strategy, see content-strategy-sms. For writing posts, see post-writer-sms.
UI/UX quality gate and build guide for every visual output — landing pages, dashboards, web apps, portfolios, and tools. ui-design remains the main entry point. Integration model: ui-design main trunk + taste-skill overlay. - ui-design owns engineering quality (track selection, component-library strategy, accessibility, responsiveness, theming, performance, preview delivery) - taste-skill is invoked inside the ui-design workflow for style decisions (Brief Inference, Design Dials, Anti-slop hard rules) MUST USE together with project-builder whenever a project produces visual HTML/CSS/JS.
Investigate an email address — MX and syntactic validation, Gravatar lookup, corporate email-format inference, breach exposure, and full mail-header analysis covering the Received chain, Message-ID and SPF, DKIM and DMARC results. Use for email OSINT, verifying whether an address exists, finding accounts registered to it, guessing a company's email format, or tracing where a suspicious message actually came from. Applies to business email compromise and invoice-fraud investigation, phishing triage, vendor-payment verification, and pre-engagement research. Reference at useosint.com/skills/what-an-email-reveals.
Benchmark CodeGraph retrieval quality on a real codebase by comparing agent behavior with vs without CodeGraph. Use when the user runs /agent-eval or asks to test, benchmark, audit, or validate a codegraph version (the local dev build or a published npm version) against a language's repo.
Comprehensive PostgreSQL database engineering skill covering indexing strategies, query optimization, performance tuning, partitioning, replication, backup and recovery, high availability, and production database management. Master advanced PostgreSQL features including MVCC, VACUUM operations, connection pooling, monitoring, and scalability patterns.
Improve academic paper writing quality for ML/CV/NLP-style papers with clear section structure, paragraph flow, and reviewer-facing presentation. Use when drafting or revising Abstract, Introduction, Related Work, Method, Experiments, or Conclusion; polishing figures/tables; checking claim-support alignment; or performing self-review before submission.
World-class data engineering skill for building scalable data pipelines, ETL/ELT systems, and data infrastructure. Expertise in Python, SQL, Spark, Airflow, dbt, Kafka, and modern data stack. Includes data modeling, pipeline orchestration, data quality, and DataOps. Use when designing data architectures, building data pipelines, optimizing data workflows, or implementing data governance.
Baidu Drive file management and document content capabilities — upload, download, save a copy, share, search files, search document passages, read full file contents, move, copy, rename, and create folders; check drive capacity, used space, and remaining space. Also supports Agent memory backup and restore (kimiclaw/maxclaw/qclaw/openclaw). Trigger when the user mentions “百度网盘/bdpan/网盘/云盘/baidu drive/Baidu Drive” and asks about file operations or drive capacity, remaining space, or usage; or mentions memory-related operations such as “备份记忆” (back up memory), “恢复记忆” (restore memory), or “查看记忆备份” (view memory backup). Do not trigger for non-file-storage operations, use of other cloud drives, local memory organization or cleanup, or PPT generation (handled separately by the baidu-wenku-aippt skill).
Build applications where agents are first-class citizens. Use this skill when designing autonomous agents, creating MCP tools, implementing self-modifying systems, or building apps where features are outcomes achieved by agents operating in a loop.
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.