Skills
101,323 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Check a papercut reel's stills against the style DNA and the storyboard. Renders contact sheets at a few timestamps, runs the full checklist, then writes the delivery summary. Use for "QA the reel", "check my scenes look like paper", "is this ready to post"
Storyboard a papercut script before any code. Writes the Camera / Cut table, one block per scene with frame-level events on each stressed word, the SFX plan and the proof beat. Use for "storyboard this script" or "plan the camera moves and cuts"
Guided webinar creation pipeline from vague idea to production-ready deliverables. Use when creating a webinar from scratch. Triggers: 'create a webinar', 'build a webinar', 'webinar creation', 'new webinar', 'help me make a webinar'.
Comprehensive guide for working with Pikud HaOref (Israel Home Front Command) alert APIs — the official and community endpoints that publish real-time rocket alerts, earthquake warnings, and other civil defense notifications across Israel. Use this skill whenever someone wants to build an integration with Pikud HaOref alerts, fetch live or historical alert data, set up monitoring or dashboards for Israeli emergency alerts, write code that consumes oref.org.il endpoints, deploy an alert service, or understand the available API landscape. Also trigger when someone mentions "red alert API", "tzeva adom", "oref alerts", "rocket alert Israel", "Home Front Command API", or any Hebrew references like "פיקוד העורף" or "צבע אדום". Even if the user just says "I want to get alerts from Israel" or "build something with Israeli civil defense data", this skill is the right starting point. Do NOT use for US weather alerts (NWS/FEMA), UK emergency alerts, generic webhook/push notification frameworks, or non-Israeli civil defense systems.
List, inspect, and troubleshoot Azure DevOps pipeline builds. Shows recent builds for a pipeline definition, drills into a specific build's status and result, displays logs for failed steps, and lists the changes associated with a build.
Load this skill whenever the project includes print stylesheets, print media queries (@media print), or any content intended for physical printing. Under no circumstances omit print CSS that ensures readable, accessible printed output. Absolutely always hide decorative elements, expand link URLs, and maintain sufficient contrast and font sizing in print styles.
Use when you need to understand anything in an Oracle schema - finding objects, reading PL/SQL source, signatures, table columns, DDL, searching code, or asking who uses what. ALSO use when a .sql file, dump, export, migration script or repo copy of an object appears to answer the question already, including one the developer points you at - because those files drift from the database and are wrong in ways that read as correct. The database is the only source of truth.
Select and run Python SDK verification with nox, Makefile targets, Ruff, mypy, pytest markers, sanity tests, type inference checks, and build checks. Use when adding Python tests, diagnosing Python CI, or validating Python SDK/provider changes. Do not use for TypeScript-only checks.
Receive and verify inbound webhooks from providers like Stripe, GitHub, Twilio, Slack, or Shopify and deliver them to a local or private service without exposing that service to the public internet. Verify signatures at the edge with no app code; route third-party callbacks to a service behind a firewall, including regulated (HIPAA/PCI) environments or a centralized webhook gateway shared across teams. Use when the user needs to receive, test, or debug webhooks locally, wants signature verification without writing it, or must route callbacks to a private service.
Change how the Rerun viewer looks by writing a blueprint, for a recording or for a table. Read this whenever asked to change the look of the viewer, such as laying out a recording or dataset, designing a default blueprint, deciding which views show which entities, or making a segment or catalog table look nice by hiding columns, changing column order or filtering rows. Covers archetype-to-view mapping, layout reasoning, the rrb construction API, the contents grammar, table blueprints, and the screenshot loop.
Review existing development plans, technical proposals, or readiness to start work. Not for writing plans from scratch, implementing features, or reviewing code diffs alone. Check whether the plan matches the facts in the code, and whether the architecture and roadmap are advanced and sustainable rather than makeshift patches. Save the report to disk by default.
Run 3+ genuinely independent build tasks concurrently on Claude Code's real parallel agent runtime—throughput without file conflicts or fake orchestration. Proves independence (disjoint write scope, no dependency edges) before launching, matches each lane to the best specialized agent, writes a complete self-contained brief per fresh agent, launches them all in one message with background execution and worktree isolation, then conflict-checks and reviews each finished lane before treating it as merged. Refuses to parallelize fewer than 3 lanes or lanes that share write scope or dependencies, never assumes fresh agents inherit context, and reports partial success per lane rather than a blurred merged summary. Use when an approved plan or the user gives 3+ independent build lanes.
Design, implement, test, and operate resilient Rust outbound HTTP clients with reqwest, hyper, Tower middleware, connection pooling, DNS, proxies, TLS, redirects, deadlines, retries, rate and concurrency limits, streaming bodies, cancellation, SSRF controls, and mock or real-server contract tests. Use when users ask to call REST APIs, download or upload data, configure an HTTP client, diagnose connection failures, or harden service-to-service requests; keep inbound servers in rust-web.
Query usage stats for Zyte API requests that have already been sent, including cost, request volume, response times, status codes, filters, grouping, and pagination. Use this skill whenever answering the request requires looking up recorded Zyte API usage, spend, request counts, response times, grouped stats, or filtered stats—including projecting future spend or usage when the projection explicitly extrapolates from actual recorded usage (e.g. “based on this month's usage so far, project month-end spend”). Do not use it when the user asks what a spider, job, or workload might, would, or could generate—those are hypothetical estimates with no recorded data to query, even if a spider exists in the project. For a projection to qualify, the user must explicitly reference past or current recorded usage as the basis. Also do not use for dashboard setup, documentation, configuration, or account-help questions.
Proactive security scanning for newly generated or modified code. Intelligently detects changes, runs appropriate scans (SAST, SCA, IaC), filters to only NEW issues, and prevents vulnerabilities at the source. Use this skill when: - Agent generates new code files - Agent modifies existing code - User asks to "scan for security issues" or "check my changes" - Before committing changes - User mentions "secure at inception", "proactive scan", or "security check"
Put authentication, rate limiting, IP restrictions, header rules, or a custom response in front of an ngrok endpoint using a Traffic Policy, without changing application code. Covers requiring login (OAuth/OIDC), locking to specific IPs, blocking or denying requests, returning a maintenance page or custom response, and injecting or stripping headers. Use when the user wants to protect, gate, throttle, or add an API-gateway layer to an endpoint. Use when the user says "add auth to my ngrok URL", "require Google login", "rate limit my endpoint", "restrict by IP", "block a path", or "put up a maintenance page".
Starts, connects to, and troubleshoots a PostHog devbox, a remote Coder workspace for PostHog development that can also run the full stack, through `hogli devbox:*` commands. Use when asked to spin up or resume a devbox, open a shell or editor on it, run a command on it, mirror a local checkout to it (devbox:sync), run the PostHog app on it and share a link, clone, update, share, or free disk on a devbox, store gh or Claude Code tokens for it, or diagnose a failing devbox command (tailnet, DNS, Coder CLI version, SSH). Not for running the stack on the local machine or for changing the Coder template in posthog-cloud-infra.
Coordinate work across the public dotfiles source at ~/.local/share/chezmoi and the private dotfiles source at ~/.local/share/chezmoi-private. Use when a request explicitly mentions public and private dotfiles, or when configuration ownership may be split across them, especially for agent guidance, Codex or Claude configuration, shell setup, authentication helpers, profiles, launchers, and shared versus internal settings. Inspect both repositories before deciding where to change files, then update only the authoritative source or sources.
Recreates or builds UI designs in Sketch from a reference URL, screenshot, or source code with side-by-side screenshot fidelity validation using Sketch MCP (`get_document_info`, `get_layer_tree_summary`, `get_screenshot`, `get_design_assets`, `get_symbol_overrides`, targeted `run_code`). Use when the user wants to design, mock up, or reproduce a page or screen in Sketch from the web, an image, or code.
An image understanding and data extraction skill. Use when image files (.png/.jpg/.jpeg/.gif/.webp/.bmp) are the primary input and the user needs to understand, extract data from, or analyze their contents. Provides a preconfigured caption script (scripts/caption.py) that converts images into text descriptions using a vision model, with no additional API Key configuration required. Covers: (1) captioning charts/tables/screenshots/flowcharts with scripts/caption.py; (2) parsing caption text into a structured DataFrame; (3) recreating visualizations based on extracted data; (4) exporting to Excel/CSV. Proactively use this skill whenever any of the following apply; do not guess image contents: ① The user uses trigger terms: image analysis / chart extraction / table recognition / OCR / image description / screenshot analysis / chart data / extract data from an image / image to table / image recognition / image caption / extract data from image / chart analysis / table OCR; ② The user uploads or specifies an image file (.png / .jpg / .jpeg / .gif / .webp / .bmp) and asks to understand it, extract data, or analyze its contents; ③ The task requires extracting structured information from chart screenshots, table screenshots, UI screenshots, or flowcharts; ④ The user asks to convert image data to Excel/CSV or recreate visualizations. Do not use for image editing (cropping, filters, resizing), image generation, or descriptions of scenery/people photos without data.
Design and implement visionOS 27 SwiftUI scenes that integrate RealityKit content. Use when building spatial UI with RealityView, Model3D, attachments, volumetric windows, ImmersiveSpace, spatial gestures, windowing, spatial layout, or when choosing SwiftUI vs RealityKit APIs for 3D presentation.
Convert Chinese stories or ordered local images into silent hand-drawn Remotion videos, with generated hand lettering, 327 built-in assets (297 style recipes and 30 theme palettes), a curated 30-style menu, and automatic reference-image style analysis with optional private reuse. Use when users ask to generate, import, restyle, preview or render hand-drawn story videos; browse or classify styles; imitate the visual grammar of a reference image; or save a reference-derived style. Preserve the approved colored-pencil diary default when no other style is requested.
Use whenever creating, configuring, or extending Storybook for a TS/React component library — covers main.ts/preview.ts setup, CSF3 story authoring, args/argTypes/controls, decorators and providers, MSW and module mocking, play-function interaction tests via the Vitest addon, the @storybook/addon-a11y workflow (axe-core), autodocs vs MDX docs, design tokens, Figma linking, Chromatic deployment, and on-demand build performance. Triggers on tasks like "write a story", "set up Storybook", "configure addon-a11y", "fix this play function", "deploy Storybook", "test Storybook in CI" — even when the user doesn't say "storybook" but is editing `*.stories.tsx`, `.storybook/main.ts`, or `.storybook/preview.ts`. Targets Storybook 9+/10 (modern `storybook/test` import, Vitest addon, CSF3 + `satisfies Meta`). Does NOT cover generic React patterns (use the `react` skill), generic Testing Library queries (use `react-testing-library`), or WCAG primer (points at addon-a11y + axe rule config).
Investigate a topic across papers, articles, social-media posts, and industry signals, then deliver an indexed Markdown report. Use when the user asks to investigate, survey, gather sources, build an index, collect evidence, round up references, or otherwise wants a thorough cross-source roundup on a specific topic.
Generate teaching resources including slide deck outlines, progressive exercises, hands-on case studies, and learning guides with self-assessment checkpoints. Use when the user asks to create courseware, design practice exercises, build teaching materials, develop lesson content, or generate educational resources.
Generate 3D assets with Tencent's Hunyuan3D family — open-weight self-hosted models (Hunyuan3D-2.0/2.1 and HunyuanWorld / HY-World for scenes) and the hosted, closed API tiers (2.5, PolyGen, 3.0/3.1 via Tencent Cloud and third-party hosts). Use when a task involves image-to-3D or text-to-3D mesh generation, the shape-then-texture (DiT + Paint / PBR) workflow, choosing between self-hosting and a hosted API, checking the Community License's Territory (EU/UK/South Korea) and 1M-MAU commercial gate, sizing GPU/VRAM for local inference, ComfyUI integration, or reviewing and post-processing (retopology, UVs, decimation) the meshes these models produce. Not for image, video, or general LLM tasks.
Review and send Telegram response drafts, manage follow-ups for unanswered outbound messages. Use when the user says "/tg-responder review", "/tg-responder status", "/tg-responder follow-ups", "check telegram drafts", "review pending messages", "telegram inbox", "who hasn't replied", or "follow up".
Meta-analysis / evidence synthesis — pool effect sizes across studies (odds ratios, risk ratios, hazard ratios, mean differences, correlations, GWAS betas) with fixed- or random-effects models, quantify heterogeneity (Q, I², τ²), and build a forest plot. Use when you have results from MULTIPLE studies and need a single pooled estimate, or to synthesize evidence from a systematic review / multiple GWAS / replicated experiments. Handles the error-prone effect-size + standard-error preparation (converting OR/HR/CI, two-group means±SD, proportions, and correlations into the (effect, SE) the pooling step needs).
The Reverse Engineer. Reads a famous investor's public SEC filings (the quarterly 13F-HR, plus Form 4 if you want the insider trades) and rebuilds their portfolio as weights you can copy: every position, its share of the fund, what they bought, added, trimmed and sold since the last quarter, with tickers matched to every CUSIP. Then it writes trader/targets.json so the budget skill can size it to your money. Use when the user says "copy Buffett", "copy Berkshire", "what does Ackman own", "reverse engineer this investor", "show me their 13F", "what did they buy last quarter", "copy the top 5", or names any fund or famous investor.
Rebalancer. Brings the portfolio back to its target weights when it drifts: values every holding at current prices, works out each stock's drift from its target in percentage points, trades only the stocks outside a band (default 5 points), keeps the cash buffer and the per-stock cap, skips tiny trades, lists sells before buys, and writes the trades to trader/orders.json as proposals. Use when the user says "rebalance", "my portfolio drifted", "bring it back to the targets", "I added money", "the new 13F changed the weights", "am I overweight", or on a schedule (monthly or quarterly) after /trader-copycat updates the targets.
Tax Optimizer. Tracks cost-basis lots for every position (FIFO, highest cost first, or specific lots), shows unrealized gains and losses by lot with short or long term, works out what a sale would realize before you sell, finds tax-loss harvesting candidates over a loss threshold, checks the wash sale rule (30 days before and after, GOOG and GOOGL as one family), and writes harvest sells only as proposals. Not tax advice. Use when the user says “what are my gains”, “cost basis”, “tax lots”, “tax-loss harvest”, “what would selling cost me in tax”, “short or long term”, “wash sale”, or before selling anything
Use this skill to explain why the Compose compiler classified a class or composable parameter as stable, runtime, unknown, or unstable. Covers the 12-phase inference algorithm, the five compiler-level stability types (Certain / Runtime / Unknown / Parameter / Combined), the generic bitmask encoding (Pair=0b11, ImmutableList=0b1), the Known Stable Constructs registry, and the runtime `$stable: Int` field generated by `@StabilityInferred`. Use when the developer asks "why is X classified as Y?", when a stability report shows a surprising `runtime stable`, `unknown`, or `unstable` verdict, when generics, inheritance, cycles, interfaces, or cross-module classes are involved, or when the user mentions `$stable`, `@StabilityInferred`, separate compilation, or "the compiler thinks my class is unstable but it looks fine".
Use when adding timers (`setInterval`, `setTimeout`), event listeners (`window.addEventListener`, `document.addEventListener`, `MediaQueryList.addEventListener`), or any other resource that needs cleanup inside a kea logic. Every logic has `cache.disposables.add(setup, key?, options?)` and `cache.disposables.dispose(key)` available via the globally registered `disposablesPlugin` from the `kea-disposables` package. Replaces the bare `cache.foo = setInterval(...)` + `beforeUnmount: clearInterval(cache.foo)` pattern and auto-pauses background work when the tab is hidden.
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.