Skills
100,928 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Create finished vertical short videos with Remotion and post them to TikTok, Instagram Reels, and YouTube Shorts. Turns a topic, article, URL, dataset, changelog, or list into a 1080x1920 kinetic-text short (large animated words, stat count-ups, top-N lists, launch and changelog videos, quotes), timed to reading speed and kept inside each app's safe zone, rendered locally with no paid API. Use when someone asks for a Remotion short, a Remotion video for TikTok/Reels/Shorts, kinetic typography or text animation video, a stats or data short, a top 5 list video, a product launch or changelog video, a quote video, faceless text shorts, or “make a short about X and post it.” Includes a review sheet with app overlays marked, a cover frame, and WoopSocial posting with confirmation.
Package a codebase into a compact, structured context bundle for an AI agent. Use when the user wants to "give the whole repo to an LLM", "pack the codebase", hits context limits sharing a project, or asks for a digest/summary of an unfamiliar repo.
Find the correct note to resolve an unresolved ingredient wikilink in a recipe note. Use when an ingredient wikilink like [[leeks]] in a recipe does not resolve to an existing note and you need to find the best matching product note to point it at. Uses islandiguana to enumerate product-tagged candidate notes, then semantically ranks the full list against the link text and its surrounding recipe ingredient line.
Genre pack: Risograph / 70s–80s print revival. Warm paper, pigment inks (one red + one blue maximum), halftone dot textures, misregistered ink offsets, rubber-stamp details, poster typography. Use for zines, events, agencies, food/drink brands, music, and anything wanting a warm analog voice. Stack on the design-taste core skill.
Evidence-based code review. Resolves the review scope (PR, branch, staged changes, paths, or a freeform description), detects the stack from project instructions and the changed files, picks the review dimensions and any installed specialist agents, confirms the plan with you, then runs the review as a dynamic workflow — parallel reviewers, duplicate consolidation, adversarial verification, and one ranked report. Use when asked to review code, a PR, a branch, staged changes, or a directory.
Use when judging a produced document PLAN — a manifest of which documents a project will produce, each with producer/tools/skills/depends_on — to decide if it is sound enough to produce from. An acceptance gate, not authoring. Judges a fourteen-condition bar single-sourced with project-document-discovery's Self-check (v6.0.0): proportional; load-bearing present; production reqs + depends_on per doc; acyclic graph; no orphan (leaf deliverables like LICENSE/README exempt); no padding; open-ended preserved; change-scoped amend delta; skills carry purpose/requirements; capability_map + product_capabilities keys valid; manifest nested with fan-out entries matching flags; all five manifest meta-sections populated (capabilities/roles/skills/tools not empty, no build-level capabilities); capability scalar on every document entry (not array). Emits exactly one terminal VERDICT: approve|revise. Review-only — not a document's internal quality, not the coherence of finished documents (reviewing-document-set).
Use for Roblox Engine API lookup during implementation: finding the correct engine class or service, confirming properties, methods, events, callbacks, datatypes, enums, globals, and built-in libraries, and verifying parameter, return, and property usage for a known engine task. Prefer this skill when the problem is referential rather than architectural.
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.
Default entry of the debug skill family when runtime observation is needed: static analysis stalled, a fix did not work, logs look correct but behavior is wrong, or fix verification needs before/after evidence. Covers escalation decision, targeted instrumentation (not shotgun logging), MRE reproduction, evidence tiers, confidence gating, escape hatches, and fix verification; composes with browser-debug-toolkit / hybrid-debug / channel enablers without demoting them. Triggers — 「打点调试」「打日志调试」「运行时调试」「运行时验证」「复现验证」「日志分析根因」「根因置信度不足」「静态分析碰壁」「修了还是不行」「日志正常但行为不对」「偶现」「调试默认入口」 / runtime evidence debugging, instrumentation debugging, escalate to runtime, evidence-driven debugging, fix didn't work, intermittent bug.
The Figma↔Storybook bridge, both directions, via the native Figma MCP. design→code: map foundation tokens (color/spacing/type) from Figma variables with design↔code parity + drift, and deliver approved Figma components (extract → build → embed), authoring stories via sb-stories' rules. code→design: build Code Connect mappings from components + stories + token parity + usage so Figma Dev Mode shows the real code. Use for 'sync my Figma tokens', 'map Figma variables to my design system', 'deliver this approved Figma design', 'check design↔code token parity', 'connect my components to Figma / code connect'. NOT for prototyping/iterating an undecided design (that's sb-explore).
Evaluating scientific evidence and claims. Covers study design hierarchy (RCT to expert opinion), effect sizes (OR, RR, NNT, Cohen's d), confounding, p-value vs clinical significance, GRADE quality assessment, reproducibility, and bias types (selection, information, confounding, reporting). Use when reading a paper or assessing claims.
Evaluate scientific claims and evidence quality. Use for assessing experimental design validity, identifying biases and confounders, applying evidence grading frameworks (GRADE, Cochrane Risk of Bias), or teaching critical analysis. Best for understanding evidence quality, identifying flaws. For formal peer review writing use peer-review.
Time-to-event modeling with scikit-survival: Cox PH (elastic net), Random Survival Forests, Boosting, SVMs for censored data. C-index, Brier, time-dependent AUC; Kaplan-Meier, Nelson-Aalen, competing risks. Pipeline/GridSearchCV compatible. Use statsmodels for frequentist, pymc for Bayesian, lifelines for parametric.
Use when the user wants Instagram research or workflow guidance for lead generation, influencer discovery, brand monitoring, competitor analysis, content analytics, trend research, or audience analysis, including profile analysis, feed collection, post or reel inspection, transcript extraction, comment analysis, reel discovery, highlight retrieval, or embed generation.
Transcribe a ScreenKite recording's mic audio, detect silence gaps in the word-level transcript, and cut those gaps from the timeline using editTimeline action=cut. Use when the user asks to "remove silences", "auto-cut dead air", "transcript-based cut", or "clean up pauses" in a .skbundle project. Do NOT use for B-roll overlay work (see use-screenkite-advanced-b-roll). Do NOT use when no .skbundle is open.
URL search param and hash state management. Use when adding or modifying URL search params, working with useSearchParams, setSearchParams, useSearchParamState, or navigate() with query strings or hash fragments, or fixing browser back/forward button issues.
Conversational security advisor that hunts for real, exploitable vulnerabilities in a codebase. Use this whenever the user asks for a security review, vulnerability scan, "find the bugs that matter", pre-ship audit, PR security pass, prompt-injection review, threat model, or mentions any specific CVE / OWASP category / auth concern / webhook concern / Stripe signature / Prisma injection / server-action safety / LLM prompt injection. Also use when the user says things like "look for security issues", "am I leaking secrets?", "is this endpoint safe?", "should I be worried about X?", or gets anxious before a production push. Grounded in Anthropic's 2025/2026 safety research (Agents Rule of Two, Sleeper Agents, prompt-injection defences) and OWASP Top 10:2025 / OWASP LLM Top 10 2025. Complements /cso - this skill is a dialog, /cso is an audit.
This skill should be used when the user asks to "perform vulnerability scanning", "scan networks for open ports", "assess web application security", "scan wireless networks", "detect malware", "check cloud security", or "evaluate system compliance". It provides comprehensive guidance on security scanning tools and methodologies.
Jimeng Seedance video creative workbench. When the user sends an image plus copy, it autonomously completes image analysis → copy expansion → camera-move matching → quality validation → API generation. Triggers: 即梦, Seedance, seedance, video generation, video prompt, AI video, camera movement, short drama, ad video, video extension, image-to-video.
Generate a demand-gen landing page for an SEO agency, complete with pain-point hook, proof, a free-audit lead magnet flow, and CTAs tuned for cold traffic. Pulls real competitive and AI-search data for the agency's target niche to make the copy specific and credible. Use when the user asks for an SEO agency landing page, lead-gen page for an SEO agency, demand-gen page, free-audit landing page, or wants to convert cold traffic into discovery calls.
Local SEO analysis covering Google Business Profile optimization, NAP consistency, citation health, review signals, local schema markup, location page quality, multi-location SEO, and industry-specific recommendations. Detects business type (brick-and-mortar, SAB, hybrid) and industry vertical (restaurant, healthcare, legal, home services, real estate, automotive). Use when user says "local SEO", "Google Business Profile", "GBP", "map pack", "local pack", "citations", "NAP consistency", "local rankings", "service area", "multi-location", or "local search".
Deep single-page SEO analysis covering on-page elements, content quality, technical meta tags, schema, images, and performance. Use when user says "analyze this page", "check page SEO", "single URL", "check this page", "page analysis", or provides a single URL for review.
Use this skill when the user asks to call an authenticated HTTP API (for example "call the GitHub/OpenAI/Slack API", "hit an endpoint that needs a bearer token") and the `sesame` CLI is already installed on this device. The agent invokes `sesame request`, which forwards the HTTP call through the user's own broker and attaches the auth header server-side. The skill does not install software, does not read credentials from the environment, and runs shell only within the fixed `sesame` subcommand surface (`request`, `status`, `hostnames`, `login`, `refresh`, `switch`, `police` — incl. its `--auto`/`--yes`/`--verify`/`--backup`/`--neutralize` migration flags — `launch`, `init`, `onboard`, `trust`, `doctor`, `proxyd`, `egress`, `proxy-key`, `help`, `secret`, `agents`, `deploy`, `update`). Skip for unauthenticated public endpoints, localhost services, or when the user has already exported a token in the environment for direct use.
Settle an effort once its pull request is approved - merge it, run the post-merge follow-ups, carry unfinished work into next-effort tickets, close the spec and tickets, then settle the session. Use when the user says to settle the effort, or when another skill says to
Terminal commands for the Sinch CLI (`sinch`, npm `@sinch/cli`). Use whenever the user runs or asks about a `sinch` command: login, Functions init/dev/deploy/logs/status, phone numbers and porting, sending SMS/WhatsApp/RCS, placing calls and configuring voice services, fax, SIP trunks, secrets and templates. Runtime code is covered by sinch-functions-node and sinch-functions-dotnet.
Build interactive presentation websites — slide decks projected to a room, controlled by one presenter. Use this skill WHENEVER the user wants to create, build, or design a presentation, slide deck, talk, or "bài thuyết trình / slide / trình chiếu", and also when they ask to add/edit slides, restyle a deck, scaffold a new presentation project, reuse their presentation setup in another repo, control slides from a phone (remote, QR code, "điều khiển slide bằng điện thoại"), or export slides to PDF — even if they don't say the word "slide" but clearly need projected talk visuals (e.g. "I'm giving a talk next week and need visuals", "dựng deck cho buổi chia sẻ", "làm mấy trang chiếu cho hội thảo"). Supports two tracks: a zero-build single HTML file, or a Vite + React project. Do NOT use it for printable documents, normal web apps, or dashboards meant for individual users to interact with.
Define and implement Service Level Indicators (SLIs) and Service Level Objectives (SLOs) with error budgets and alerting. Use when establishing reliability targets, implementing SRE practices, or measuring service performance.
Use when the user wants to interact with LinkedIn or X (Twitter): outreach, posting, audience insights, message triage, account research, comments/reactions, companies, company Page management, jobs, bookmarks, connected-account management, billing audits, bug reports, and feature requests. Triggers include "search LinkedIn", "find <name> on LinkedIn", "look up this tweet", "my X bookmarks", "show my home timeline", "check my messages", "from:<username>", "report a bug", "request a feature", "send feedback", "let's get started with social", "set me up", "log me in", "connect my LinkedIn/X", and explicit `/social`. Operates the `social` CLI (npm `@usesocial/cli`); never call LinkedIn's or X's HTTP APIs directly.
Start here for Deepgram speech-to-text. Use when a task says “speech to text,” “STT,” “transcribe,” “transcription,” “live transcription,” “captions,” “diarization,” “nova-3,” “Flux,” “turn detection,” or “end of turn.” Picks the model family (Nova on /v1/listen for general transcription, Flux STT on /v2/listen for conversational audio with built-in turn detection), gets a first request working, and routes to the api, docs, recipes, starters, examples, and per-language SDK skills for everything else
Python statistical modeling: regression (OLS, WLS, GLM), discrete (Logit, Poisson, NegBin), time series (ARIMA, SARIMAX, VAR), with rigorous inference, diagnostics, and hypothesis tests. Use scikit-learn for ML; statistical-analysis for test choice.
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.