MLOps
模型部署、评估与运维
funsloth-upload
chrisvoncsefalvay
Generate comprehensive model cards and upload fine-tuned models to Hugging Face Hub with professional documentation
Healthsafety
robdtaylor
UK health and safety management for manufacturing operations. Covers risk assessments, COSHH, PPE, incident reporting, LOTO, manual handling, and emergency procedures. Aligned with UK HSE regulations and IATF 16949 clause 7.1.4. USE WHEN user says 'health and safety', 'risk assessment', 'COSHH', 'PPE', 'accident', 'incident', 'LOTO', 'lockout tagout', 'manual handling', 'emergency procedure', or needs safety compliance guidance.
funsloth-train
chrisvoncsefalvay
Generate Unsloth training notebooks and scripts. Use when the user wants to create a training notebook, configure fine-tuning parameters, or set up SFT/DPO/GRPO training.
reinforcement-learning
pluginagentmarketplace
Q-learning, DQN, PPO, A3C, policy gradient methods, multi-agent systems, and Gym environments. Use for training agents, game AI, robotics, or decision-making systems.
funsloth-hfjobs
chrisvoncsefalvay
Training manager for Hugging Face Jobs - launch fine-tuning on HF cloud GPUs with optional WandB monitoring
funsloth-local
chrisvoncsefalvay
Training manager for local GPU training - validate CUDA, manage GPU selection, monitor progress, handle checkpoints
hytale-custom-blocks
MnkyArts
Create custom block types for Hytale with textures, physics, states, farming, and interactions. Use when asked to "add a custom block", "create a new block type", "make blocks farmable", "add block interactions", or "configure block physics".
mlops-deployment
pluginagentmarketplace
Docker, Kubernetes, CI/CD, model monitoring, and cloud platforms. Use for deploying ML models to production, setting up pipelines, or infrastructure.
openscad-iterative-modeling
fboldo
Iteratively generate OpenSCAD (SCAD) code, render PNG previews, visually critique the result, and refine in a tight loop until the model matches. Use when the user asks to create a 3D model, design a part, make a bracket/enclosure/mount, render in OpenSCAD, export STL, or any request that can be expressed as parametric CSG geometry. Requires the OpenSCAD MCP server tools (render_scad_png, export_scad_stl).
@tank/typescript
tankpkg
"Advanced TypeScript patterns for type-safe applications. Triggers: typescript, TS, type, interface, generic, union, discriminated union, branded type, utility type, conditional type, mapped type, template literal, infer, satisfies, as const, tsconfig, strict mode, type guard, type narrowing, type-safe."
concept-explorer
Semprini
Use this skill when the user asks "what is [concept]", "how does [feature] work", "explain [section]", "compare MD-DDL to [tool]", "why does MD-DDL [design choice]", or any question about a specific part of the standard. Also use when the user wants to understand the reasoning behind a design decision or the trade-offs between alternatives (entity vs enum, canonical vs bounded context, etc.).
langchain-models
evanfang0054
Initialize and use LangChain chat models - includes provider selection (OpenAI, Anthropic, Google), model configuration, and invocation patterns
@tank/vector-graphics-gen
tankpkg
Generate vector graphics (SVG) for websites and apps using AI APIs. Primary platforms: QuiverAI Arrow (highest quality, #1 SVG Arena) and fal.ai with Recraft V4 for native SVG generation, plus image-to-SVG conversion. Covers text-to-vector generation, image vectorization, prompt engineering for clean vector output, SVG optimization (SVGO), and web integration patterns (React, Vue, sprites, dark mode). Synthesizes QuiverAI documentation, fal.ai documentation, Recraft AI documentation, SVGO docs, and SVG specification. Trigger phrases: "vector graphic", "generate SVG", "SVG icon", "vector illustration", "AI vector", "fal.ai", "FAL_KEY", "recraft", "recraft v4", "text to SVG", "image to SVG", "quiver", "quiverai", "arrow model", "vectorize image", "SVG generation", "vector logo", "AI illustration", "icon generation", "SVG for web", "vector art API", "fal-ai/recraft", "generate icon", "web illustration", "SVG background", "vector pattern", "clean SVG", "production SVG", "text-to-vector"
langchain-multimodal
evanfang0054
Work with multimodal inputs/outputs in LangChain - includes images, audio, video, content blocks, and vision capabilities
@tank/auth-patterns
tankpkg
Authentication and authorization patterns for any language or framework. Covers JWT internals (structure, algorithms, attacks, validation), OAuth2 grant types (Authorization Code, PKCE, Client Credentials, Device Code), session management (cookies, expiry, fixation, distributed), RBAC/ABAC/ReBAC (role modeling, authorization policies, Zanzibar), OpenID Connect and social login (ID tokens, account linking, provider patterns), MFA (TOTP, WebAuthn/passkeys, backup codes, step-up auth), and authentication security (XSS/CSRF, token storage, credential stuffing, rate limiting). Synthesizes RFC 6749, RFC 7519, RFC 6238, W3C WebAuthn Level 2, NIST SP 800-63B, and OWASP Authentication/CSRF cheat sheets. Trigger phrases: "JWT", "OAuth2", "OAuth 2.0", "session management", "RBAC", "ABAC", "role-based access", "authorization model", "OpenID Connect", "OIDC", "social login", "MFA", "multi-factor authentication", "TOTP", "WebAuthn", "passkeys", "refresh token", "access token", "PKCE", "auth flow", "implement authentication", "implement auth", "sign in with", "cookie security", "HttpOnly", "SameSite", "token storage", "XSS auth", "CSRF protection", "credential stuffing", "account linking", "backup codes", "permission system", "login security", "password hashing"
jupyter-notebooks
Kang-chen
Jupyter notebook operations: editing cells, reading notebooks, executing, and format conversion. Trigger when working with .ipynb files for: (1) Creating/editing/deleting/reordering cells, (2) Reading notebook content, (3) Executing notebooks with papermill, (4) Converting to HTML/PDF/script formats. Supports Cursor EditNotebook tool, Jupytext workflows, and nbformat.
eld-model
CAPHTECH
Deprecated: /eld-spec ã使ç¨ãã¦ãã ããã ãã®ã¹ãã«ã¯å¾æ¹äºæã®ããã®ã¨ã¤ãªã¢ã¹ã§ãããã¹ã¦ã®å¼ã³åºã㯠/eld-spec ã«è»¢éããã¾ãã
orientation
Semprini
Use this skill on first contact, when the user asks "what is MD-DDL", "where do I start", "what can I do here", or describes their role and goals. Also use when the user asks for a general overview of the standard, the agent ecosystem, or the workflow.
apple-foundation-models
jakerains
Build Apple Intelligence features with Foundation Models and Image Playground on iOS, iPadOS, macOS, and visionOS. Use when implementing or debugging SystemLanguageModel, LanguageModelSession, guided generation with @Generable, tool calling, custom adapters, model availability checks, context-window limits, local on-device inference, routing to larger-model paths (App Intents Use Model or custom backend fallback), and ImagePlayground or ImageCreator APIs.
@tank/opencode-agent-creator
tankpkg
Create specialized OpenCode agents that assume specific roles — frontend architect, DevOps SRE, database specialist, code reviewer, or any domain expert. Covers agent anatomy (markdown and JSON formats), role design (persona, expertise, behavioral directives), system prompt engineering, tool permissions, oh-my-opencode integration (categories, skill injection, delegation wiring), and converting existing Tank skills into standalone agents. Synthesizes OpenCode source (sst/opencode), oh-my-opencode v3.x agent system, and production agent analysis. Trigger phrases: "create agent", "opencode agent", "new agent", "convert skill to agent", "make an agent", "agent from skill", "custom agent", "specialized agent", "agent template", "agent for", "agent configuration", "agent role", "create a specialist", "build agent", ".opencode/agent"
langchain-chat-models
evanfang0054
使用 LangChain 中的聊天模型集成指南，包括 OpenAI、Anthropic、Google、Azure 和 Bedrock
eld-model-law-card
CAPHTECH
Deprecated: /eld-spec-card law を使用してください。 このスキルは後方互換のためのエイリアスです。すべての呼び出しは /eld-spec-card law に転送されます。
langchain-structured-output
evanfang0054
使用 Zod 模式、类型安全响应和自动验证从 LangChain 代理和模型获取结构化的验证输出
sam3
jakerains
Create and work with Meta SAM 3 (facebookresearch/sam3) for open-vocabulary image and video segmentation with text, point, box, and mask prompts. Use when setting up SAM3 environments, requesting Hugging Face checkpoint access, generating inference scripts, integrating SAM3 into Python apps, fine-tuning with sam3/train configs, running SA-Co or custom evaluations, or debugging CUDA/checkpoint/prompt pipeline issues.