Skills
101,323 packaged capabilities and playbooks for agents. Install any of them with one command.
All
When the user wants to establish or update their developer audience context. Also use when starting any other developer marketing skill to ensure foundational context is loaded. Trigger phrases include "developer persona," "target developers," "who are our developers," "developer profile," "ICP," or "ideal customer profile."
Receive and verify Discord webhook events. Use when setting up Discord webhook handlers, debugging Ed25519 signature verification, handling PING endpoint validation, or processing events like APPLICATION_AUTHORIZED, ENTITLEMENT_CREATE, or LOBBY_MESSAGE_CREATE.
Create a Business Requirements Document (BRD) - Layer 1 of the SDD flow, defining business needs, objectives, and success criteria. Use when starting a new project or feature. Single-document authoring primitive; for end-to-end or batch generation the autopilot (`doc-brd-autopilot`) drives this skill.
Use this skill when the user needs SHA-1, SHA-256, or SHA-512 digest acceleration on BlueField-2 silicon via doca_sha (absent on BF-3)—selecting one-shot `doca_sha_task_hash` versus incremental `doca_sha_task_partial_hash` for chunked input, consulting `doca_sha_cap_*` for per-algorithm digest-size limits and input alignment, or decoding DOCA_ERROR_* statuses such as BAD_STATE on a resumed partial hash. Also load it for implied asks like "accelerate SHA-256 on the NIC", "streaming digest over chunks", "digest output size mismatch", or "is SHA-512 present on this card". Refuse and route to the proper owner for cryptographic-hash theory (collision resistance, SHA-3 selection), the AES-GCM / Compress / DMA libraries, or SDK install and BFB bring-up.
Use this skill when the user is doing hands-on DOCA Telemetry Exporter programming on a host where DOCA is installed — defining a doca_telemetry_exporter_schema and event types, creating sources, picking a publish surface (typed events / opaque events / the metrics counter-gauge-histogram API / OTLP logs / NetFlow), walking the schema-then-source lifecycle, or debugging DOCA_ERROR_* failures from the exporter API. Trigger even when the user does not explicitly mention "DOCA Telemetry Exporter" or "doca_telemetry_exporter_*" — typical implicit phrasings include "publishing counters from my DOCA app", "BAD_STATE when I report an event", "consumer/DTS sees nothing but my report succeeded", "how do I export NetFlow/IPFIX records", or "should I link the exporter or the telemetry service". Refuse and route elsewhere for the receiving DOCA Telemetry Service (DTS), plain stdout logging via doca_log, or real-time event subscription back into the app via doca-comch — those belong to other skills.
Changing DOCA versions safely: upgrading or downgrading the host DOCA packages, refreshing the BlueField BFB, bumping the NGC DOCA container tag, or rolling back, with the discipline detect -> report -> ASK -> only then a guided upgrade. Trigger on "is there a newer DOCA", "I want the newest DOCA features, can I just apt upgrade in place", "should I move to the next release", "the upgrade stopped midway and dpkg is in a broken state", "my component is being deprecated, what now", or "roll me back", even without the word upgrade. Detect what is installed, find the newer release, report the gap, then STOP and ask for explicit confirmation; never upgrade automatically. Do not use it for a runtime error on an unchanged install (doca-debug). Version detection alone belongs to doca-version, first-time install to doca-setup, any hardware / firmware / reboot step to doca-hardware-safety, and public-docs or sunset routing to doca-public-knowledge-map.
Generate meaning-carrying editorial data-illustrations in the monotykamary / Linear aesthetic (near-black grayscale, Inter display + mono labels, hairline framed figures) with a single coral accent. This is a GENERATIVE GUIDE, not a template gallery: it teaches the "claim -> geometry" method so any session can invent the RIGHT figure for what a post actually argues, then render it as a self-contained, theme-aware, reduced-motion-safe HTML/SVG figure. Use when writing/illustrating a blog post, essay, spec, or slide and you want a chart, diagram, or illustration that carries a thesis (not decoration) in the house style. Triggers: "make a chart/diagram/illustration for this", "visualize this argument", "monotykamary/Linear style figure", "editorial graphic".
Close a pull-request review session — summarize what was found, decided and changed, for the original author who was not in the room, and check that every change made during the review is documented with its reason before the summary is posted. Then runs end-acceptance to settle the acceptance lessons. Use when the user says the PR review is finished, asks to wrap up or write up a review, or invokes /end-pr-review <pr or branch>. Recommends a verdict; never merges and never pushes unasked.
Feishu calendar, tasks, approvals, attendance, and OKR: check availability and shared free time; book meeting rooms; create, edit, delete, and reply to events (including participants, recurrence range, sharing, and organizer transfer); manage tasks, subtasks, groups, and lists; initiate, withdraw, approve, reject, transfer, roll back, add approvers, remind, or CC on approvals; check clock-ins and attendance statistics; view OKR cycles, create or update objectives and key results, and report progress and comments. Use when the user mentions events, meeting rooms, to-do tasks, approvals, clock-ins, leave, or OKR. Not for contact lookup or schema/raw API calls to unwrapped interfaces (feishu-cli-platform), email (feishu-cli-mail), past meetings/recordings/minutes (feishu-cli-meetings), or sending notifications or listening for real-time approval events (feishu-cli-messaging).
Fetch presentation slides from a URL and convert them to PDF. Supports direct PDF URLs, PPTX URLs, OneDrive sharing links, and RevealJS HTML presentations. For RevealJS, also extracts slide text content and links to a markdown file. USE FOR: download slides, fetch slides from URL, convert PPTX to PDF, get RevealJS slides, OneDrive slides.
Build or repair adaptive Flutter layouts across devices, constraints, breakpoints, overlap, clipping, orientation, and input modes. Use with flutter-accessibility when text scaling causes reflow, clipping, or disappearing controls; not for general widget state architecture.
Build and evaluate low-latency spoken, multimodal, and translation experiences with Google Gemini Live API and Gemini Enterprise Agent Platform Live API. Use when a media-production or voice-agent workflow needs real-time audio input/output, barge-in, voice configuration, live transcription, live translation, tool/function calling during a spoken session, WebSocket session design, latency QA, quotas/cost review, or Google/Vertex data-governance tradeoffs.
Use when generating Weaverse project export JSON for import into the Weaverse editor. Triggers on requests to create, build, or produce a Weaverse project JSON file from a site migration, design spec, page layout description, or existing export. Also use when converting section plans into importable project and pages data.
Scaffold new Go projects and services: directory structure, cmd/ and internal/ conventions, when to use a flat layout, module naming, and main package wiring. Use when: "new Go project", "scaffold a service", "create project structure", "start a Go module", "how do I organize a new service", "set up folder structure". Not for: reviewing an existing architecture (go-architecture-review), DI wiring (go-dependency-injection), CI setup (go-ci).
Use Google Lyria music generation models through Google Cloud / Vertex AI / Gemini Enterprise Agent Platform for production music beds, songs, vocal tracks, image-conditioned music, short-form social audio, ad music, and video scoring. Covers Lyria 2, Lyria 3 Clip, and Lyria 3 Pro model selection, prompt construction, rights/privacy checks, SynthID/C2PA provenance, artifact custody, and QA.
Hardens CockroachDB user privileges by auditing and tightening role-based access control, reducing admin grants, restricting PUBLIC role permissions, and applying least-privilege principles. Use when reducing excessive privileges, cleaning up admin access, or implementing RBAC best practices.
Design a teacher think-aloud that models historical thinking strategies with a specific document. Use when planning explicit strategy instruction or when students follow a protocol without understanding the underlying reasoning.
Control native macOS apps that have no API and leave reproducible evidence: capture a specific window (even when occluded in the background), click and type, probe whether an app can be automated, control Chromium-embedded apps via CDP with zero focus, run product testing and multi-model batch runs. For browser work use huashu-chrome; for browsing old screenshots use disk-keeper.
Guides competitive idea generation and ranking using tree-structured search (up to N_I=21 candidates across technique/domain/formulation axes) and Elo tournaments (4 dimensions: novelty, feasibility, relevance, clarity). Produces a ranked direction summary and full research proposal. Use when: user has a research direction and needs concrete ranked ideas, wants to compare multiple approaches, or mentions 'rank ideas', 'compare approaches', 'which idea is best', 'research proposal'. Do NOT use for finding a research direction from scratch (use research-ideation) or planning the paper itself (use paper-planning).
Proactively design, improve, or review high-quality Dear ImGui tools, C++ editor panels, game-development interfaces, debugging dashboards, visualization screens, and desktop workflows. Use when Dear ImGui is named or supplied code contains direct evidence such as `ImGui::`, `imgui.h`, `ImGuiContext`, or ImGui backend symbols. For an unlabelled C++ editor/tooling request, use only when the user explicitly asks whether Dear ImGui is suitable; otherwise keep framework choice open. Do not trigger for a task that explicitly names another UI framework such as Qt, GTK, WPF, wxWidgets, JUCE, or Slate. Cover UI architecture, information hierarchy, dockspace layouts, design systems, custom components, UX states, keyboard/gamepad interaction, DPI/fonts, performance, and code quality.
Analyze an already-collected infrahub-collect diagnostic bundle — traceback and failure triage across service logs, incident correlation, and known-issue matching against opsmill/infrahub GitHub issues. TRIGGER when: the user says they collected a diagnostic bundle, gives you a bundle path, or pastes bundle contents — and asks what's wrong, wants the logs or tracebacks analyzed, asks "what does this bundle say", wants errors triaged/correlated, or asks whether a crash is a known issue. DO NOT TRIGGER when: no bundle exists yet — collect one first (use infrahub-collecting-diagnostics), filing the issue itself (use infrahub-reporting-issues), or querying live data (use infrahub-analyzing-data)
Use when the context window fills too fast, per-turn input token cost is too high, or the aggregate input token bill needs explaining and fixing. Two audits: (1) overhead — tool schemas, skills index, system fixed text, user profile, memory recall measured live, ranked, reduced; (2) cost accounting — total input token volume across sessions, cron fires, and tool traffic, with cache-aware interpretation, pre-checked config levers, and scheduled delta checks. scripts/audit_overheads.py for per-turn overhead, scripts/audit.py for totals.
Scrape Instagram profiles, posts, reels and comments as JSON through the Bright Data Scraper API. Use when the user asks for any Instagram data - an account's followers or bio, a post's likes and comments, the recent posts or reels of an account, the comments on a post - or says "instagram scraper", "instagram api", "instagram data". Routes between the CLI (`bdata pipelines run` with the Instagram dataset ids), the Python SDK (`client.search.instagram.*` and `client.scrape.instagram.*`), and the MCP scraper tools. Carries the credit costs, the reels timeout, the `post_type` values that work, and the error-row cases the API returns as success.
When the user wants to optimize inventory levels, calculate safety stock, determine reorder points, or minimize inventory costs. Also use when the user mentions "inventory management," "safety stock," "EOQ," "reorder point," "service level," "stockout prevention," "ABC analysis," "inventory turns," or "working capital reduction." For warehouse slotting, see warehouse-slotting-optimization. For multi-echelon systems, see multi-echelon-inventory.
Processing, transforming, or moving datasets that may exceed RAM on a single low-compute box — covers memory discipline (streaming, generators, dtype shrinkage), I/O access patterns (sequential vs random, mmap, async), data formats (Parquet vs CSV vs JSON, predicate pushdown), chunking & batching, spill-to-disk (external merge sort, DuckDB/Polars), pipelining (bounded queues, backpressure, checkpointing), codec selection (zstd/lz4/gzip), concurrency for I/O-bound workloads (asyncio, threads, prefetch), and observability (iowait vs CPU%, rows/sec, py-spy/strace). Trigger on "process a large file", "stream this", "out-of-core", "OOM kill", "this is slow", or code with `pd.read_csv` of multi-GB files, `requests.get(...).content` on big bodies, `BytesIO` on unbounded inputs, per-row INSERTs, sequential `requests.get` loops, falling `tqdm` rates — even if I/O or memory isn't mentioned. Complement to computer-science-algorithms.
Legal document drafting -- contracts, memos, briefs, complaints, demand letters, opinions, discovery, settlements, ToS, privacy policies. Full pipeline: document structure, per-section writing, Bluebook citation, case law lookup (CourtListener API), regulation lookup (eCFR API), DOCX output, and TDD-style verification (defined terms, cross-references, placeholders, boilerplate, citation format). Triggers on: 'draft a contract', 'write a legal memo', 'create an NDA', 'write a brief', 'legal document about', 'draft a complaint', 'terms of service', 'privacy policy', 'demand letter', 'settlement agreement', 'legal opinion', 'discovery requests', any request to produce a legal or law-related document.
Filter Etsy products by multiple criteria using the EHunt MCP tool `_ehunt_productQuery` (display name: “Etsy Product Query”), including keywords/URLs, price, sales, favorites, reviews, listing date, category, handmade/vintage types, and Pick/Bestsell/Raving, among others. Trigger when users mention EHunt Etsy products, Etsy listings, Etsy product research, Etsy bestsellers, Etsy handmade, Etsy vintage, ehunt items, Etsy Product Query, or `_ehunt_productQuery`. Also trigger this skill when users search for products on Etsy, check sales, prices, or tags, or filter products, even if they do not mention EHunt.
General-purpose image generation execution layer supporting three image-generation APIs: MuleRun Nano Banana 2, APImart GPT Image 2, and Atlas Cloud GPT Image 2. Switch with --provider. Supports generation (text-to-image) and edit (image editing with a reference image) modes, with single-image and batch execution. This is an infrastructure skill called by other skills, not intended for end users. When another skill (such as cover-hero or inline-diagram) needs to call an API to generate images, use this skill's scripts/generate.py. Do not use for prompt writing, style injection, or business validation—those are handled by the calling skill
Use this skill to plan, direct, troubleshoot, and hand off provider-independent live or hybrid livestream event productions, including run of show, crew, capture, audio, graphics, switching, guests, encoding, ingest, transport choices, latency, redundancy, accessibility, moderation, monitoring, incidents, recording, and delivery.
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.
