Skills
99,448 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Facilitates the learning step of a proven customer-interview method — the recording half: turning raw material from ONE customer conversation (transcript, notes, memory dump) into a brief per-person debrief file, mapped against the numbered interview-question list (Q1, Q2, … with H-number tags). One concise answer per question actually asked, key phrases kept verbatim, unasked questions honestly skipped, a one-line commentary only where in-the-moment context helps later analysis, and an addenda section for everything heard that fit no question. Load when the user has just done a customer interview and says 'here's the transcript,' 'process my call notes,' 'file this conversation against my questions,' or 'add this interview to the record.' Do NOT load for writing goals, hypotheses, or interview questions (earlier steps), for synthesizing patterns across multiple interviews and updating hypotheses (the next step), or for summarizing meetings or transcripts unrelated to customer interviews.
Plan and schedule social media and content marketing calendars. Monthly and weekly planning, content mix ratios, theme days, platform-specific timing, and batch creation workflows. Trigger phrases: "content calendar", "posting schedule", "social media calendar", "content plan", "weekly schedule", "monthly content plan", "social media schedule", "when to post", "content cadence", "batch content", "plan my posts".
Maps a customer journey across stages, touchpoints, emotional curve, pain points, and moments of truth into a markdown artifact with an optional mermaid timeline or flowchart. Use when synthesizing existing research into the shape of a customer's experience, end-to-end or for one phase. Without research signal yet, run discover-interview-synthesis or measure-survey-analysis first; refuses to fabricate emotional or behavioral data.
OpenAlgo indicator expert. Use when the user asks about technical indicators, charting, plotting indicators, creating custom indicators, building dashboards, real-time feeds, scanning stocks, indicator combinations, or using openalgo.ta. Also triggers for indicator functions (sma, ema, rsi, macd, supertrend, bollinger, atr, adx, ichimoku, stochastic, obv, vwap, crossover, crossunder, exrem).
Design and implement integration architectures connecting financial systems — APIs, FIX protocol, ISO 20022, event-driven patterns, batch feeds, idempotency, and resilience. Use when building custodian integration pipelines, implementing FIX connectivity for order routing, designing ISO 20022 or SWIFT migration messaging, building batch file processing for custodian feeds or EOD reconciliation, implementing idempotency for transaction APIs, designing retry or circuit breaker patterns, mapping data between systems with different schemas, or troubleshooting integration failures causing recon breaks. Trigger on: FIX protocol, ISO 20022, custodian feed, batch processing, API design, idempotency, circuit breaker, dead letter queue, data mapping, integration architecture, SWIFT migration, mTLS, file feed, event-driven, message broker.
Linux kernel module skill for writing and debugging loadable kernel modules. Use when writing LKMs with Kbuild, adding module parameters, creating /proc and sysfs entries, implementing character devices, debugging with KGDB or ftrace, or handling module signing for Secure Boot. Activates on queries about Linux kernel modules, loadable modules, Kbuild, module parameters, /proc filesystem, sysfs, character devices, KGDB, or module signing.
Draft and refine prompts for video generation models (including text-to-video, image/keyframe-to-video, and reference-driven generation), and create character-sheet prompts for image models when the goal is character consistency before image-to-video. Use when a user asks for a "video prompt", a model-specific prompt such as MiniMax H3, Seedance 2.0, Seedance 2.5, Ovi, Veo 3, Wan 2.2, Wan Animate 2, LTX-2, LTX-2.3, or LTX-2.5, or a consistent-character prompt such as "character sheet prompt", "character turnaround", "character reference sheet", or "photographic identity sheet".
Embed compliance controls into account opening and verify regulatory readiness. Use when designing CIP/KYC identity verification gates, implementing OFAC and sanctions screening at onboarding, collecting beneficial ownership certification for entity or trust accounts, building risk-based approval tiers that route applications by risk level, defining compliance screening requirements and exception tracking, adding senior investor protections (FINRA Rules 2165/4512) or trusted contact procedures, establishing CDD risk ratings and ongoing monitoring triggers, or preparing account opening procedures for SEC or FINRA examination. For the operational pipeline these controls plug into, see account-opening-workflow.
Facilitates the first step of a proven customer-interview method: deciding exactly what you're trying to learn, written as numbered goal questions (G1, G2, …) that your hypotheses and interview questions will later be designed to answer. Interviews the user about their business — new idea or established company, B2B or B2C — then drafts a tailored goal-question list, critiques and revises it with them, and preserves the result in a GOALS.md file. Load when the user wants to interview customers, plan customer discovery or customer development, validate a startup idea, or answer unaskable questions like what to charge, how to position, who the ideal customer is, or what to build. Do NOT load for job or hiring interviews, for writing survey questionnaires, for analyzing customer interviews that have already been conducted, or when a goal-question list already exists and the user wants the next step (hypotheses).
Review GitBook change requests from Claude Code by calling the GitBook REST API directly with curl (no CLI) — the reviewer-side companion to cr-create (the authoring side over the same API). Discover the change requests that need review (filter by who opened them, by space, or across a whole org), get the GitBook app link to review the diff, summarize what actually changed in a CR, then leave comments and optionally submit a review verdict (approve / request changes). Use this whenever someone wants to review docs change requests over the raw API (curl/HTTP), asks "what CRs are open / waiting on me / opened by <person>", "show me the change requests in <space>/<org>", "summarize what changed in this CR", "review this change request", "leave a comment on a CR", or "approve / request changes on a CR". For the authoring side (create a CR, push content, request reviewers, fix comments) over the API, use cr-create instead.
EasyEDA Pro API skill for AI agents. Use when working with the EasyEDA Pro EDA software, including PCB design, schematic editing, footprint/symbol management, and project operations. Supports live debugging in EasyEDA and EasyEDA extension development. Provides a complete API reference (120+ classes, 62 enums, 70 interfaces), extension development documentation, and a WebSocket bridge server that executes code in the running EasyEDA Pro client. Trigger on: "嘉立创EDA，启动！", "立创EDA，启动！", "EDA，启动！", "EasyEDA", "PCB", "schematic", "footprint", "EDA", "circuit board", "嘉立创EDA", "原理图", "PCB设计". **IMPORTANT**: 嘉立创EDA's English name is **EasyEDA**. They are the SAME product. Never use other transliterations like "EasyEDA Pro" (unless specifically versioned), "EASYEDA", "easyeda", etc. Always use "EasyEDA" for English references and "嘉立创EDA" for Chinese references.
Implement, fix, review, migrate, test, or debug Drift persistence in Flutter apps using SQLite, drift_flutter, type-safe Dart queries, generated tables, StreamBuilder or Riverpod StreamProvider UI, write operations, transactions, schema migrations, web assets, isolate sharing, and local database testing. Use when a Flutter task mentions drift, local database storage, SQLite, reactive database streams, CRUD, schemaVersion, build_runner, drift_dev make-migrations, migration tests, or Flutter-specific database setup across mobile, web, or desktop.
Python best practices for writing production-grade code. This skill should be used when writing, reviewing, or refactoring Python code. Triggers on tasks involving Python development, error handling patterns, dictionary operations, and code quality improvements.
Use this skill to orient an agent to the architecture of the RevenueCat Android SDK (10.x). Covers how RevenueCat reorganizes BillingClient, the Google Play Developer API, and RTDN into a single SDK, and how Products, Packages, Offerings, Entitlements, and CustomerInfo connect.
Use this skill immediately when the user mentions merge conflicts that need to be resolved. Do not attempt to resolve conflicts directly - invoke this skill first. This skill specializes in providing a structured framework for merging imports, tests, lock files (regeneration), configuration files, and handling deleted-but-modified files with backup and analysis.
Implement the Syncfusion React 3D Chart component from the @syncfusion/ej2-react-charts package. Use this skill when users need 3D column, bar, stacked column, stacked bar, or 100% stacked chart variations. Covers Chart3DComponent, axis configuration (category, numeric, datetime, logarithmic), data binding, multiple panes, data labels, legends, tooltips, selection, print/export, theming, and accessibility.
Create responsive navigation headers with the AppBar component. Build top, bottom, or sticky app bars with buttons, menus, icons, and sidebars. Use when building navigation elements, toolbars, action bars, or responsive headers with dynamic content layout. Includes color modes (Light, Dark, Primary, Inherit), size modes (Regular, Prominent, Dense), spacing, separators, and full accessibility support.
Exhaustively extract UX patterns from a reference web app. Walks every screen, captures screenshots of every state, records interaction patterns, copy verbatim, keyboard shortcuts, responsive treatments, motion, and empty/error/loading states. Produces a reusable pattern library that other audits can compare against. The inverse of ux-audit — asks 'what is the bar?' rather than 'does this match the bar?'. Trigger with 'learn from X', 'extract patterns from X', 'study X's UX', 'reverse engineer the UX of X', 'build a pattern library from X'.
The user shared a video URL, a YouTube/TikTok/stream link, a local video file, a screen recording, a meeting recording, or a playlist/folder of videos — “watch this”, “summarize this video”, “what's in this recording”. Use this to actually watch the video — download, extract frames, OCR, transcribe, and index it — instead of guessing from the title or asking the user to describe it.
Process account maintenance requests across the account lifecycle. Use when changing a client address or contact info with identity verification, updating beneficiary designations after marriage, divorce, birth, or death, re-registering or re-titling an account to a trust or new entity, selecting tax lot methods or fixing cost basis records, applying legal or compliance holds or Reg T freezes, setting up systematic withdrawals or standing instructions, processing a death notification and estate account setup, handling a QDRO, power of attorney, or guardianship, closing accounts and managing escheatment, or designing data quality review programs.
Facilitates the second step of a proven ideal-customer (ICP) method: distilling raw company observations into the few deep-truth attributes that matter, then classifying each as a strength, a weakness, or deliberately both. Takes an observations list (O1, O2, … — file or pasted), proposes attributes one at a time with their supporting observations, kills generic ones with the Opposite Test ('we love our customers' dies), classifies each with a concrete rubric (a third of the market sees it that way, or some customers buy/refuse for it alone), and records the result in STRENGTHS-WEAKNESSES.md (S1, S2, … / W1, W2, …). Load when the user has raw observations and wants to distill them, asks 'what are our real strengths and weaknesses,' or wants to classify what they learned from the honest-look exercise. Do NOT load to gather the raw observations (the previous step), to derive keystones or deal-breakers from classified attributes (later steps), or for a person's individual strengths and weaknesses.
Facilitates the final step of a proven customer-interview method: distilling everything a round of interviews produced (GOALS.md, HYPOTHESES.md, QUESTIONS.md, and a directory of per-interview debriefs) into a single FINAL-REPORT.md the whole company can use. Top: a summary as brief as possible without losing salient information. Below: numbered findings (F1, F2, …) tagged validated / disproved / directional / watch / untested, every one citing debriefs and quoting customers verbatim, plus per-area briefs that marshal the evidence for ideal-customer definition, positioning, pricing, marketing & sales, and product priorities. Load when the user says 'write up what we found from the interviews,' 'summarize the interview results for the team,' or 'turn the interviews into a report.' Do NOT load for updating hypotheses from interviews (the synthesis step), for recording one conversation (the debrief step), or for actually doing the positioning, ideal-customer, or pricing work the report feeds.
Guide for querying databases through DBHub MCP server. Use this skill whenever you need to explore database schemas, inspect tables, or run SQL queries via DBHub's MCP tools (search_objects, execute_sql, and the opt-in explain_sql and health_check). Activates on any database query task, schema exploration, data retrieval, or SQL execution through MCP — even if the user just says "check the database" or "find me some data." This skill ensures you follow the correct explore-first workflow instead of guessing table structures.
Route generative media requests before any creative planning or provider execution. Use this when the user asks to generate, modify, dub, animate, or assemble image, video, audio, workflow, or analysis-derived media and the first decision is which generation controller should own the job.
Drive Graphify from its CLI to build, refresh, query, export, and serve a durable code/corpus knowledge graph. Use when the user wants `.graphify/GRAPH_REPORT.md`, `graph.json`, `graph.html`, `graphify update`/`summary`/`query`/`path`/`explain`/`tree`, change-aware review context, git-hook or watch-based refresh, a stdio MCP graph server, or an install into jeo, jeopi, gjc, opencode, Claude, Codex, or Gemini. Also covers the honest structural fallback when native extraction is empty or misleading. Route simple locate/reference work to `codebase-search`, narrative knowledge-base work to `llm-wiki`, and project-memory handoff to `opencontext`. Triggers on: graphify, graphify update, graphify query, knowledge graph CLI, GRAPH_REPORT.md, graph.json, codebase graph, graph refresh, graphify install, graphify serve, review context, affected flows.
Copy learned patterns from another local project, or browse, download and publish patterns through the IPFS pattern marketplace. Use when a new project should start with patterns an older one already learned, or when sharing patterns with a peer.
Evaluate ecommerce price candidates using unit economics, historical observations, elasticity analysis, scenario modeling, and controlled experiments. Use when a seller asks what price to test, how price changes could affect contribution or revenue, how to estimate elasticity, how to optimize a bundle or tier, or how to design a price experiment across Amazon, Shopify, TikTok Shop, Walmart, eBay, or other channels. Do not claim a proven optimal price without sufficient clean data, and do not change live prices without explicit authorization.
Measure and raise straight-through processing (STP) rates in securities operations through zero-touch, exception-based processing. Use when measuring STP rates and analyzing manual touchpoints in an existing process, replacing review-all workflows with exception-based processing, evaluating RPA vs API-based vs hybrid automation for legacy systems, building exception queuing, categorization, and auto-resolution workflows, conducting process mining or root cause analysis on exception volumes, or setting STP rate targets and continuous improvement programs. For approval chains, four-eyes controls, and SLA monitoring, see workflow-automation.
First-run setup for [email protected] — installs ONNX/Brain/SONA add-ons, registers the MCP server, and verifies the install via `doctor`
A thin orchestration layer for embeddedskills: discovers projects in the current workspace, selects a build/flash/debug/observe backend, chains .embeddedskills/state.json, and aggregates the results of the underlying skills. Triggers when the user explicitly enters one of the commands: "一键构建烧录", "自动诊断", "串起 build -> flash -> debug -> observe", or explicitly invokes /workflow.
Facilitates the second step of a proven customer-interview method: recording the user's current best guesses — hypotheses — as numbered, falsifiable statements (H1, H2, …), each mapped to the goal questions it addresses, so interviews can confirm or contradict them instead of confirmation bias quietly filtering what's heard. Takes a GOALS.md goal-question list as input (file or pasted), elicits what the user believes goal by goal, sharpens vague beliefs into testable claims, prunes to hypotheses whose resolution would actually change what the user builds, targets, charges, or says, and preserves the result in HYPOTHESES.md. Load when the user has goal questions and wants to write hypotheses, list their assumptions, or record predictions before interviewing customers — 'I have my goals, what's next,' 'help me write down what I believe about my customers.' Do NOT load for writing the interview questions themselves, for analyzing interviews already conducted, or for statistical hypothesis testing.
Facilitates the learning step of a proven customer-interview method — the synthesis half: reads a directory of per-interview debrief files against the working HYPOTHESES.md and QUESTIONS.md and proposes evidence-cited updates ONE at a time — double down, tune numbers, mark disproved, park heard-once observations in a 'That's funny' watch section, add new hypotheses with new questions — applying each agreed change directly to the files with a change log, and ending with a stop-or-continue-or-re-aim verdict on the interviewing itself. Load when the user says 'what did we learn from these interviews,' 'update my hypotheses from the interviews,' 'synthesize my interview notes,' or 'should I keep interviewing?' Do NOT load for recording a single conversation into a debrief (the previous step), for writing initial goals, hypotheses, or interview questions (earlier steps), or for deciding what to build next (comes after this method).
Facilitates the third step of a proven customer-interview method: translating hypotheses into open-ended, unbiased interview questions — each a miniature experiment designed to test one hypothesis without leading the interviewee. Two modes: given a single hypothesis, it grills the question into shape and outputs the final question in the chat; given a HYPOTHESES.md file, it iterates the whole list, grouping related hypotheses, and maintains a QUESTIONS.md file (numbered Q1, Q2, … mapped to H-numbers) as a live, resumable artifact. Loads when the user has hypotheses and wants interview questions, asks how to phrase a question for customers without biasing the answer, or says 'turn my hypotheses into questions' or 'help me ask about X without leading.' Do NOT load for writing goal questions or hypotheses (earlier steps), for conducting or analyzing the interviews themselves, for survey/questionnaire design, or for job interviews.
Guide which client disclosure documents must exist, what each must contain, and when and how they must be delivered to clients of investment advisers and broker-dealers. Use when the user asks about Form ADV Part 2A or 2B content, Form CRS content and delivery, prospectus delivery obligations, privacy notice delivery, trade confirmation timing, account statement distribution, electronic vs paper delivery compliance, or the content and client-delivery side of the annual brochure update. Also trigger when users mention 'onboarding document checklist', 'what disclosures do we owe new clients', 'brochure supplement for a new adviser', or 'CRS conversation starters'. (For filing mechanics and regulatory deadlines — IARD amendments, Form PF, 13F — use regulatory-reporting; for whether fee disclosure content is adequate, use fee-disclosure.)
Design, generate, and deliver client performance reports across all channels, covering quarterly reports, tax reporting, portal integration, and compliance review. Use when the user asks about building or redesigning report templates, choosing what to include in quarterly or annual client reports, transitioning from print to digital delivery, integrating a client portal, presenting net-of-fee performance with benchmarks, managing report production timelines, or handling e-delivery consent. Also trigger when users mention 'client reporting', 'quarterly performance report', 'report customization', 'tax lot report', '1099 supplement', 'report disclaimers', 'UHNW reporting', or 'report QA process'.
Upgrades Google Cloud Armor security policies from OWASP CRS 3.0 and 3.3 to OWASP CRS 4.22. Preserves signature opt-in and opt-out lists, translates evaluatePreconfiguredExpr to native evaluatePreconfiguredWaf dictionary parameters, remaps signature IDs to v042200, drops retired signature IDs, and updates structured preconfiguredWafConfig fine-tuning exclusions. Evaluates operational risk (High Breaking Change, Medium Attention Required, Low Clean Upgrade), including Method Enforcement breaking changes (PUT/PATCH/DELETE blocking) and sensitivity coverage expansions. Supports both in-place rule upgrades and smooth preview mode deployments with priority spacing and a 200-rule quota guardrail. Use when a user or agent needs to audit, upgrade, or transition Cloud Armor WAF rules to CRS 4.22. Don't use for non-WAF Cloud Armor rules (IP rate-limiting, geo-blocking, reCAPTCHA) or AWS/Azure WAF migrations.
Drive an end-to-end GitBook docs review flow from Claude Code by calling the GitBook REST API directly with curl (no CLI) — create a change request, push content (update an existing page and create a new page), request reviewers, notify Slack, then pull review comments back in, fix them, re-push, and resolve. This is the authoring-side companion to cr-review (the reviewer side over the same API). Use this whenever someone wants to run a "docs review in GitBook" loop from the terminal/agent against the raw API (curl/HTTP), mentions creating a change request via the API, pushing content into a CR, "pull in the latest comments and fix them", requesting review on docs, or showing engineers how to collaborate on GitBook docs from Claude + Slack without a CLI.
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.