Claude Code プラグイン
公開されている Claude Code marketplace からの 6,118 件のプラグインです。それぞれにスキル、スラッシュコマンド、エージェント、フック、MCP サーバーがまとめてインストールされます。
公開されている Claude Code marketplace からの 6,118 件のプラグインです。それぞれにスキル、スラッシュコマンド、エージェント、フック、MCP サーバーがまとめてインストールされます。
.claude-plugin/marketplace.json を含むリポジトリを提出すると、すべての plugin を掲載します。
marketplace を提出するКонтроль качества 1С-разработки: стандарты, принципы ООП и паттерны, XML метаданных, блокирующие гейты и проверяемый след прогона.
Review and maintain reusable Terraform modules against a written profile of the module family's rules: pull request review with one verdict, findings-only review of a local change, and a maintainer skill that makes the change.
developmentUse when you want a delegated second opinion or implementation from GPT (Codex), Gemini, Grok (xAI), or OpenRouter (config-driven, 400+ models) - seven expert subagents (Architect, Plan Reviewer, Scope Analyst, Code Reviewer, Security Analyst, Researcher, Debugger) and bundled ask-gpt/ask-gemini/ask-grok/ask-openrouter/ask-all/consensus commands, advisory (read-only) or implementation (write; Grok and OpenRouter are advisory-only).
developmentUse when writing, reviewing, or debugging Terraform/OpenTofu modules, tests, CI, scans, or state ops - diagnoses failure mode (identity churn, secrets, blast radius, CI drift, state corruption) with version-aware guards.
developmentOffers to set up the Hedgehog build discipline in any project you open
Turn one brief into three complete, comparable design directions (A/B/C) inside your project. Edit one region with snapshots and undo, then integrate the chosen direction with a rollback path. Needs Node.js 22 or later.
Search the web, read the clean text of a page by URL, crawl a few pages of one site and search Telegram channels through the NeuralDeep Search API with an sk- key. A standard-library Python client takes the key from --api-key, NEURALDEEP_SEARCH_API_KEY or NEURALDEEP_API_KEY, prints compact text or the raw JSON and exits with meaningful codes.
researchUse when the user asks to create, draw, render, or update a technical diagram, architecture scheme, process flow, lifecycle, or system map from supplied information and expects editable source plus a finished image.
workflowTest Open Graph, Twitter Card and social-link meta tags fully offline: feed it a URL (localhost works) or an HTML file and get a report with mocked preview cards for Telegram, X/Twitter, Facebook, Discord, Slack, WhatsApp, LinkedIn, Reddit and iMessage plus rule-violation findings. No third-party API.
workflowCross-review code with several console code agents at once: Claude Code, Codex, Coddy, Cursor Agent, Devin, OpenCode and others review the same brief in parallel and blind to each other, then the orchestrating agent verifies every finding against the code and decides alone what matters and what to fix. Linux, macOS and Windows; reviewers chosen once per agent, for one project or for all of them.
workflowDrive a remote Windows machine over RDP from an AI agent: the RDP client runs on an isolated Xvfb display, the operator watches the same screen in a VNC window and can take over by hand, the session survives between agent turns, and lock screens are recovered by an NLA reconnect instead of typing the password.
opsCheck whether a host or a whole namespace of a fleet harness repository is healthy, read-only: a POSIX pass over boot history, clock, units, disks, OOM, containers, resolvers, ports, NVIDIA and Intel Arc GPUs and PCIe AER, drift against the host card, stacks against the mirror, and functional probes of every service, rolled up into healthy, warming up, degraded, broken or unreachable.
opsOnboard a server into a fleet harness repository: reach it over ssh with a confirmed host key, discover it read-only with server-discovery, place it in a namespace, write the host card and README, mirror its compose stacks and system configs, add it to the namespace Topology, regenerate the index, and hand a box with an incomplete GPU stack to gpu-server-setup once the user agrees.
opsRead-only discovery for a fleet harness repository: a POSIX probe sent over ssh collects OS, kernel, CPU, RAM, disks, Docker, NVIDIA and Intel Arc GPUs, addresses, ports, tunnels and DNAT rules; discover.py turns it into the host card after comparing it with the card, render_index.py validates every card and regenerates the host index and an optional self-contained network map.
opsPrepare a Linux server with NVIDIA or Intel Arc GPUs for LLM and neural-network workloads: NVIDIA driver + CUDA from the official repo or the in-tree Intel i915/xe driver with distro firmware, Docker Engine with verified GPU access (NVIDIA Container Toolkit or /dev/dri passthrough) as the mandatory base, staged verification gates for both vendors, optional docker-compose presets for vLLM, Infinity, OpenWebUI and Ollama plus Intel Arc stacks (llama.cpp SYCL/Vulkan, Ollama, OpenVINO, vLLM), GPU power limits persisted via systemd, case fans driven by GPU temperature and Intel Arc idle-power fixes (ASPM).
opsCapture clean, repeatable screenshots of a GUI desktop app (IDEs, browsers, native apps) on an isolated Xvfb display: drive it with python-Xlib, capture with ImageMagick, annotate with a highlight box, crop or arrow. Never touches the operator's live desktop.
opsGenerate secure MikroTik RouterOS v7 configs from plain language, or review and harden an existing /export. Hardening baseline, four presets, per-model device reference, offline validator.
networkingEstimate the floor (self-) cost of a self-hosted LLM token: electricity + hardware amortization, measured on the server. Not the market API price.
opsClassical formal logic from G. Chelpanov's textbook: concepts, judgments, syllogisms, induction, fallacies, laws of thought.
reasoningFix a bug with a reproduction test first, then the fix, then the full suite, then a short report. Requires a clear bug description.
workflow