Plugins

from ehartye/sf-browser-control

Salesforce-focused compound engineering plugin (V3 skills-first architecture) with parallel agent dispatch, institutional knowledge compounding, and Apex/LWC/Flow/Integration coverage.

Base skills and rules for working in Salesforce DX projects.

Skill for buying, selling, and managing GPU compute on SF Compute from Claude.

Agent skills for Singapore government tech projects — platform integrations, compliance, accessibility, and delivery workflows.

SGLang operator reference suite — HiCache (three-tier hierarchical prefix cache: GPU HBM → host DRAM → distributed L3 via Mooncake/3FS/NIXL/AIBrix/EIC/SiMM/file/LMCache) and the SGLang Model Gateway (`sgl-model-gateway`, formerly `sgl-router`) — Rust router fronting vLLM/SGLang inference workers on Kubernetes with cache-aware/prefix-hash/consistent-hash routing, K8s label-selector service discovery, PD-disaggregation, and `--enable-mesh` CRDT sync between gateway replicas. Pairs with the `vllm` suite for vLLM-side reference and `inference-cache` for engine-agnostic KV-cache transport.

Benchmark a **running SGLang-XPU server** on an Intel GPU using `sglang.bench_serving`. Measures TTFT, TPOT, ITL, end-to-end latency, and throughput against the OpenAI-compatible endpoint. Use after sglang-xpu-run. Not for vLLM servers (use vllm-xpu-bench) or no-server PyTorch (use torch-xpu-bench).

Serve a Hugging Face safetensors model on an Intel GPU using SGLang's XPU backend with the OpenAI-compatible API. Covers pulling the pre-built `intel/sglang-dev:latest` image, fixing the render-group and UMD/kernel compatibility issues that affect non-root sglang images, the SYCL_UR / Level Zero env vars needed on Battlemage, the `--device xpu --attention-backend intel_xpu` flag set, multimodal serving, and how to validate output content (not just HTTP 200). Use when the user needs SGLang's RadixAttention prefix caching or grammar-constrained output; for broad-coverage serving on Intel today prefer vllm-xpu-run, and for benchmarking a running server use sglang-xpu-bench.

Skills for using the shablon CLI to render Jinja2 templates from a .shablon/ directory into a project tree.

shadcn/ui — die vollständige Bibliothek: alle ~60 Komponenten (Base-UI & Radix-UI-Variante, kompletter Quellcode, Props/Anatomy, alle Examples), 100+ Blocks, 70+ Charts, Theming/Colors/Tailwind-v4, Dark-Mode, Installation (Next/Vite/Astro/Remix/Laravel/TanStack/React-Router), CLI, components.json, eigene Registry bauen, MCP-Server, Forms, RTL, monorepo. Liefert den shadcn-MCP-Server mit.

Browse, search, and install shadcn/ui components via the shadcn MCP server, plus project-aware knowledge of shadcn/ui patterns, theming, and registry authoring

shadcn/ui registry MCP.

shadcn/ui component patterns: composition over custom builds, semantic theming tokens, proper form architecture, accessibility, and CLI workflow.

from Dieshen/claude_marketplace

from nthplusio/functional-claude

shadcn-vue — der Vue-Port von shadcn/ui (auf Basis von reka-ui, Tailwind v4): alle 64 Komponenten mit komplettem Vue-Quellcode, Props/Slots/Emits, allen Demos und Reka-UI-API, dazu Blocks, Charts, Theming/Colors/Dark-Mode, Installation (Vite/Nuxt/Astro/Laravel/SvelteKit u.a.), CLI, components.json, eigene Registry, Forms (vee-validate/Zod), MCP-Server. Liefert den shadcn-vue-MCP-Server mit.

Sentinel shadow reviewer — second independent reviewer on every PR. Focuses on security + correctness.

Plugin from ZCZHAO-1999/shakespeare-writing-studio

ShanireZ's personal C++14 style for competitive-programming and online-judge solutions.

ShanireZ's personal C++14 style for competitive-programming and online-judge solutions.

SHAP 기반 모델 해석성 분석을 수행합니다. Summary Plot, Waterfall Plot, Force Plot 등을 생성합니다.

Plugin from abrahamFerga/plenipo-agents

Plan a feature by reading the codebase first, deciding everything the repo can already answer, and asking only what is expensive to reverse — data models, public surfaces, vocabulary, scope, UI shape, business rules. Batched multiple-choice questions, then a written plan. Ships with `domain`, which maintains DOMAIN.md and docs/adr/ so those answers outlive the plan. Does not enter plan mode and does not implement.

Focus a runaway blog draft — find the spine, cut or split tangents, crystallize sentences one at a time. Never writes for you.

Shape Up project workflow — strategy, frame, shape, bet, kickoff, scope, breadboard, and dependency-map appetite-driven development cycles.

share — publish an HTML artifact (a plan, viz, or report) to a shareable link on your own Cloudflare R2 (zero egress, ~$0) via `agents artifacts share`. `/share` posts a public link with an auto-generated OG cover (a screenshot of the page's hero) so it unfurls into a preview card in Slack/iMessage/Twitter/Discord; `/share --private` posts an unlisted, auto-expiring (`--no-cover --expire 7d`) link with no preview card (unguessable but still public-read — not authenticated). Requires a one-time `agents artifacts setup` (provisions an R2 bucket + Worker) or `agents artifacts share join`.

Put an agent-authored page in front of a person beyond the terminal and hear back: pick a direction, approve a quote, mark up a mockup, a living board, or just show something. One link, structured answer back.

Publish HTML or Markdown documents to share.eregistrations.dev and get shareable URLs. Commands: /share <file>, /share list.

from susuyan/skills

Publish prepared share-kit posts to Xiaohongshu through Chrome with preview, verification, and duplicate protection