Skills
100,928 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Check that a form actually rejects bad input: the error message renders, the submit button stays disabled, and no request goes out. Use when validation logic was written but never driven, when a "required" or pattern check looks right on screen but was never proven, or when a bug report says the form submitted invalid data anyway.
Skill for removing hardcoded video subtitles/watermarks. Automatically sets up an environment based on YaoFANGUK/video-subtitle-remover and performs subtitle removal. Triggered when the user asks to "remove video subtitles", "remove watermarks" or "get rid of this video's subtitles".
Turn a video file someone already has (an MP4, a Loom download, a long call or screen recording) into Jams (shareable jam.dev recordings with a transcript), one per issue, cut where the speaker moves to a new problem, each titled and filed in a folder. Use when the user says "upload these to jam", "turn this recording into jams", "split this video by bug", "make these jams instead of mp4s", or has a recording too long for one Jam.
Predict RNA secondary structure, MFE folding, base-pair probabilities, RNA-RNA interactions via ViennaRNA Python bindings. Pipeline: sequence → MFE → partition function and pair-probability matrix → dot-bracket → duplex. Use for siRNA/sgRNA targeting, ribozyme design, RNA accessibility. Use RNAfold CLI for batch use without Python.
Publish posts to your VK (ВКонтакте) profile or community wall and read your own recent posts, via the official VK API (wall.post / wall.get). Use when the user wants to post to VK, cross-post an article for a Russian-speaking audience, or list their own recent VK wall posts with engagement. Auth uses a VK access token (community access key recommended).
Use this skill for safe, read-only queries of VMware infrastructure — no destructive operations exist in the codebase; a test enforces it. Directly handles: a one-glance cross-cluster health summary, object-centered VM/host/datastore investigation drill-downs (correlating surrounding infrastructure + recent events), a cross-vCenter "what needs attention now?" rollup, list VMs/hosts/datastores/clusters, active alarms, recent events, VM details. Always use vmware-monitor when the user asks to "list VMs", "check vSphere alarms", "show host status", "is anything on fire", "what needs attention now", "what is happening around this VM/host/datastore", "investigate this VM" — or needs read-only VMware info before making changes. Do NOT use for any write operations — this skill is read-only and has no code path that creates, modifies, or deletes a vSphere resource. For VM modifications use vmware-aiops, for networking use vmware-nsx, for metrics/capacity use vmware-aria. For load balancing/AVI/AKO use vmware-avi.
Use when working with Outpost Bio's open microbiome foundation models - the Waypoint checkpoints (Waypoint-6m, Waypoint-45m, Waypoint-170m), the Atlas pretraining corpus, the Compass eight-task benchmark, or the `waypoint` CLI from the `waypoint-bio` package. Covers embedding microbiome samples, fine-tuning on taxonomic abundance data, benchmarking a checkpoint on Compass, pretraining a GPT-2 model on taxonomic abundance profiles, and converting MetaPhlAn, Kraken2, QIIME 2, or MGnify abundance tables into waypoint format.
Control Webflow Analyze and Optimize from JavaScript via the Browser API. Use when managing tracking consent, integrating CMPs (OneTrust, TrustArc, CookieYes), tracking experiment variations, setting custom visitor attributes, or personalizing user experiences.
Capture screenshots of live websites and web pages via API. Use when users need to screenshot a URL, capture website previews, archive web pages, generate thumbnails of sites, or document web content. Ideal for monitoring, archiving, sharing website previews, or creating visual references of web pages.
Analyse consequential choices with outside views, incentives, inversion, scenario arithmetic, opportunity costs, and explicit update conditions. Use for decision memos, premortems, strategy choices, or a requested red-team; not routine factual questions or a reason to overanalyse a small decision.
High-velocity solo development workflow. Idea to production same-day. 10 commands: plan, spike, ship, fix, review, spec-review, focus, done, drop, workflow. Auto-activates on: “plan,” “spec,” “ship,” “spike,” “fix,” “debug,” “repair,” “spec-review,” “review spec,” “analyze spec,” “challenge spec,” “focus,” “what should i do,” “prioritize,” “overwhelmed,” “what should i work on,” “done,” “finish,” “complete,” “drop,” “abandon,” “workflow,” “what's next,” “whats next,” “next step,” “what now.”
End-to-end test-fix workflow generate test sessions with progressive layers (L0-L3), then execute iterative fix cycles until pass rate >= 95%. Combines test-fix-gen and test-cycle-execute into a unified pipeline. Triggers on "workflow:test-fix-cycle".
WordPress migration and upgrade review. Use when reviewing database migrations, option versioning, upgrade routines, data backfills, dbDelta usage, release compatibility, rollback risk, or when user mentions "migration review", "upgrade routine", "dbDelta", "plugin upgrade", "data migration", "version bump", "backfill", "schema change", or "backwards compatibility". Detects unsafe migrations, missing version guards, destructive data changes, and upgrade-flow risks in WordPress code.
Use when writing or modifying Python code that imports `xolars` — pairing an xarray Dataset with per-dimension Polars frames, constructing the `Xolars` container, selecting with isel/sel, or doing zarr+parquet round-trips via write/open. Skip for plain xarray or polars work that doesn't touch the `Xolars` class.
Research X data with Xquik, the world's fastest and cheapest X Twitter scraper service with the most complete X data. Use only when the request or active context identifies X or Twitter. Use for tweet search, tweet lookup, user discovery, profile timelines, threads, followers, trends, exports, or monitoring plans. Do not use for generic posts, other networks, or local dataset analysis that needs no X retrieval. Keep reads bounded. Require explicit confirmation before private reads, writes, persistent resources, or bulk jobs. Not affiliated with X Corp.
Use when writing or reviewing SQL, schema definitions, or application code that targets YugabyteDB's PostgreSQL-compatible YSQL API (port 5433). Triggers on CREATE TABLE, indexes, connections, transactions, sharding, migrations from PostgreSQL, or any mention of YugabyteDB with SQL.
Estimates intracellular metabolic fluxes from steady-state carbon-13 isotope-tracing measurements using validated atom maps, mfapy isotope simulation, constrained multistart fitting, and flux-profile diagnostics. Use for 13C-MFA, carbon tracing, mass isotopomer distributions (MDVs/MIDs), positional isotopomers, parallel tracer experiments, and determining whether labeling data constrain a pathway flux. Distinguishes measured-label inference from COBRA flux balance analysis and flags experiments requiring nonstationary MFA.
Technical SEO for AI search engines. Checks AI crawler accessibility, llms.txt compliance, robots.txt configuration, server-side rendering. Outputs problems + fixes (generates llms.txt, robots.txt modifications). Use when user says "AI crawlers", "llms.txt", "GPTBot", "GEO technical".
Connect a Yandex KIT store to the current AI client, or reconnect it by replacing its token. Use for natural-language setup requests, including Russian text or voice transcripts such as «Связь с магазином», «Подключим магазин», «Переподключим магазин», «Переустановим связь с магазином» and «Поменяем токен». Treat short «Связь с магазином» as an orientation request.
Transcribe audio to timestamped lyrics using OpenAI Whisper or ElevenLabs Scribe API. Outputs LRC, SRT, or JSON with word-level timestamps. Use when users want to transcribe songs, generate LRC files, or extract lyrics with timestamps from audio.
API + Python SDK for ordering cell-free protein expression and binding assays. Submit sequences for expression (10–100 µg), measure binding affinity (KD) against targets, track status, and retrieve results programmatically — no wet-lab setup. Built for ML-guided directed evolution and antibody/nanobody optimization. Requires Adaptyv account and API key.
Create or redesign React and Tailwind sections, pages, flows, shells, and components. Use when the user wants new UI or a redesign. Implement one requested or selected direction directly, or generate distinct alternatives when the user asks to compare options.
Use ado-axi to work with Azure DevOps — work items, pull requests, pipelines, repositories, and raw REST calls — through token-efficient TOON output. Use when the task involves Azure DevOps, ADO, boards, work items, WIQL, pull requests, code reviews, build pipelines, or Azure Repos.
Add a custom Python tool to an AG2 `Agent` using the `@tool` decorator. Use when the user wants to give an Agent a new capability backed by Python code (API calls, DB queries, computations, file ops). Covers sync and async tools, parameter typing, Pydantic schema customisation, returning typed `Input` / `ToolResult` (text / data / images / binary), `final=True` early-exit, and dependency injection via `Context` / `Inject` / `Variable` / `Depends`.
Pause an AG2 `Agent` mid-run to collect human input via `context.input()`, or gate a tool call with `approval_required()` middleware. Use when the user wants the agent to ask for confirmation, request missing info (passwords, API keys, data), or have a human approve sensitive / irreversible / expensive tool calls (sending emails, deleting records, payments).
Persist agent state across runs, shape what the LLM sees per turn, and cap history to fit a context window. Covers `KnowledgeStore` (memory / sqlite / disk / redis), `KnowledgeConfig` (`store=`, `expose_tool=`, `write_event_log=`, `compact=`, `aggregate=`, `bootstrap=`) and its opt-out flags, aggregation strategies (`WorkingMemoryAggregate` with `prompt=` override, `ConversationSummaryAggregate`), assembly policies (`WorkingMemoryPolicy`, `EpisodicMemoryPolicy`, `ConversationPolicy`, `SlidingWindowPolicy`, `TokenBudgetPolicy`, `AlertPolicy`), compaction (`TailWindowCompact`, `SummarizeCompact`), and the lifecycle events (`AggregationStarted/Failed`, `CompactionStarted/Failed`, `EventLogFailed`). Use when the user wants the agent to remember between conversations, manage long histories, or control prompt assembly.
Single-agent recursion and parallel fan-out within one AG2 `Agent` — auto-injected `run_subtask` / `run_subtasks(parallel=True)` (opt in via `tasks=TaskConfig(...)`) for self-delegation, and `Agent.as_tool()` as a lightweight no-hub way to call one named agent from inside another. Use when a single coordinator wants to break work into its own sub-tasks, fan out concurrent sub-tasks, or invoke a specialist agent as a tool. Covers context flow, recursion safety, and `persistent_stream` for sub-task history. **For two or more agents actually collaborating with a registry, durable channels, governance, or turn-taking, use `ag2-network-quickstart` instead** — the network is the standard multi-agent pattern in AG2.
Use Arena for heterogeneous independent design and cross-review, second opinions, code/bug/plan reviews, and evidence-based analysis. Cues: 用 arena, 讨论一下, 审核一下, 交叉审核, 第二意见, 完全体 arena, 伴随子代理, 边做边检查, companion review. Discussions start with independent proposals; a named draft/diff is reviewed directly. Bare 继续 resumes an unfinished Arena checkpoint. Supports supervised execution and persistent concurrent companion checks. Not for trivial lookups or formatting; an enabled companion still accompanies ordinary work.
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to “open a website”, “fill out a form”, “click a button”, “take a screenshot”, “scrape data from a page”, “test this web app”, “login to a site”, “automate browser actions”, or any task requiring programmatic web interaction.
Manually recall durable knowledge from an ai-memory instance via the MCP (semantic/hybrid search, recent pages, prose catch-up). Use when the user asks to search the memory, 'have we done/decided X', 'what's the state of Y', or 'catch me up' — and you want an explicit recall rather than waiting for the auto-handoff.
Creates gallery-quality algorithmic art using p5.js with seeded randomness, interactive parameter controls, and seed navigation, outputting a single self-contained HTML artifact with Anthropic branding. Use when the user requests generative art, algorithmic art, flow fields, particle systems, or code-based visual art using p5.js.
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.