MLOps
模型部署、评估与运维
local-ai-agents
microsoft
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the privacy/cost/offline trade-offs. Based on Lesson 17 of AI Agents for Beginners. USE FOR: run an agent locally, offline agent, on-device agent, Foundry Local, Qwen function calling, local tool calling, local RAG, Chroma vector database, local MCP server, privacy-preserving agent, hybrid local and cloud agent, small language model agent, engineering assistant on my machine. DO NOT USE FOR: deploying agents to the cloud at scale (use deploying-scalable-agents / Lesson 16), building your first agent concept (Lesson 01), Foundry (cloud) hosted agents, GPU cluster / server-side inference provisioning.
train-robot-loop
HangYu8123
Discover and run the authoritative controller-style robot policy train, evaluate, and improve flow. Use when asked to train a robot until a metric target is met, keep improving a policy, or run a long autonomous training loop.
skill-marketplace
williamwg2025
技能市场，发现、安装、管理 OpenClaw 技能。支持评分、评论、排行榜。
tinybrain
javimosch
Train tiny neural networks in pure machin (MFL) and ship them as JSON artifacts any MFL app embeds — MLP + SGD backprop for labeled data (classifiers/regressors, incl. bag-of-words intent routing), neuroevolution (GA + fitness closure) for control tasks (game AI), a deterministic PRNG for reproducible runs, and an agent-first CLI (train/predict/eval/guide). Use when an MFL program needs a learned controller, classifier, or scorer — NOT for generative text.
wandb-experiment-memory
wilfred-dore
Query Weights & Biases experiment history to retrieve past optimization runs, compare configurations, and inform the next optimization decision. Use this skill when asked to review past AI optimization experiments, find the best configuration tried so far, understand what has already been explored, or suggest the next experiment to run based on W&B history.
ecommerce-images
shuliuzhenhua-sys
电商商品图生成工作流技能。接收用户提供的商品原图，按模式生成主图、详情图或两者；详情图按套生成并在执行前询问用户需要几张；默认调用 banana-proxy，失败时回退到 baoyu-image-gen；仅支持用中文风格名选择主图/详情图风格。
strict-json-output-hardening
Haruk1y
Improve strict JSON generation reliability for fine-tuned language models using parser-based evaluation, prompt alignment, and targeted retraining loops. Use when outputs are malformed, have extra text, or violate schema/range constraints.
grant-thinking-cn-biology
martellevaliant19
Use when evaluating biology grant ideas in the Chinese funding context (NSFC, MOST, etc.) — diagnosing project legitimacy, mechanism-centered scientific questions, reviewer-aware logic, innovation discipline, feasibility, and scope control across funding levels (youth, general, key).
senior-data-engineer
nimeshgurung
World-class data engineering skill for building scalable data pipelines, ETL/ELT systems, and data infrastructure. Expertise in Python, SQL, Spark, Airflow, dbt, Kafka, and modern data stack. Includes data modeling, pipeline orchestration, data quality, and DataOps. Use when designing data architectures, building data pipelines, optimizing data workflows, or implementing data governance.
nlm-to-editable-pptx
Laggom
Use when the user has an image-based / non-editable PPTX or PDF (especially exported from NotebookLM, or any "scanned" deck where the slides are full-page pictures with no selectable text) and wants it converted into an editable PowerPoint with real, editable text boxes. Trigger this whenever the user mentions NotebookLM slides, "make this PPT editable", "the text isn't selectable", "convert this PDF deck to editable slides", Korean slide OCR, or rebuilding a deck so the text can be edited — even if they don't say the word "skill". Works on Windows/macOS/Linux and is portable to Databricks.
cohere-streaming
RSHVR
Cohere streaming reference for real-time text generation, tool use events, and RAG citations. Covers all stream event types and async streaming patterns.
video-gen
maxgent-ai
AI video generation with text-to-video, image-to-video, and first/last frame control. Use when users ask to generate or create videos from text prompts or images.
V-JEPA 2 Self-Supervised Training
sovr610
This skill should be used when the user asks to "train V-JEPA model", "implement JEPA pretext task", "set up EMA target encoder", "configure self-supervised training", "implement smooth L1 loss", "create training loop for V-JEPA", "optimizer configuration", "learning rate schedule", "warmup cosine decay", "EMA momentum schedule", "collapse prevention", "predictor architecture", "masked prediction loss", "DROID fine-tuning loop", "annealing phase", "cooldown training", or needs guidance on V-JEPA 2 self-supervised learning, training loops, predictor design, or optimization strategies.
Hugging Face
imoonkey
```
fine-tuning-serving-openpi
majiayu000
Fine-tune and serve Physical Intelligence OpenPI models (pi0, pi0-fast, pi0.5) using JAX or PyTorch backends for robot policy inference across ALOHA, DROID, and LIBERO environments. Use when adapting pi0 models to custom datasets, converting JAX checkpoints to PyTorch, running policy inference servers, or debugging norm stats and GPU memory issues.
Visualization & Interpretability
sovr610
This skill should be used when the user asks to "visualize activations", "plot spike rasters", "show attention heatmaps", "visualize workspace", "plot reasoning traces", "create t-SNE embeddings", "interpret model decisions", "visualize HTM patterns", "show neuromodulator levels", "plot training curves", "create activation maps", "visualize feature maps", "explain predictions", "debug model behavior visually", or needs guidance on visualization, interpretability, or explainability tooling for the brain_ai system.
senior-ml-engineer
nimeshgurung
World-class ML engineering skill for productionizing ML models, MLOps, and building scalable ML systems. Expertise in PyTorch, TensorFlow, model deployment, feature stores, model monitoring, and ML infrastructure. Includes LLM integration, fine-tuning, RAG systems, and agentic AI. Use when deploying ML models, building ML platforms, implementing MLOps, or integrating LLMs into production systems.
groq-api
diskd-ai
Groq API integration for building AI-powered applications with ultra-fast LLM inference. Use when working with Groq's Chat Completions API, Python SDK (groq), TypeScript SDK (groq-sdk), tool use/function calling, vision/image processing, audio transcription with Whisper, streaming responses, text-to-speech, content moderation with Llama Guard, batch processing, or any Groq API integration task. Triggers on mentions of Groq, GroqCloud, or fast LLM inference needs.
agentic-development
campbellmcgregor
Conversational guidance for building software with AI agents, covering workflows, tool selection, prompt strategies, parallel agent management, and best practices based on real-world high-volume agentic development experience. Use this skill when users ask about setting up agentic workflows, choosing models, optimizing prompts, managing parallel agents, or improving agent output quality.
feedback-mastery
VisualxIntelligence
Navigate difficult conversations and deliver constructive feedback using structured frameworks. Covers the Preparation-Delivery-Follow-up model and Situation-Behavior-Impact (SBI) feedback technique. Use when preparing for difficult conversations, giving feedback, or managing conflicts.
rails-testing-rspec
shivamsinghchahar
Write Rails tests with RSpec, FactoryBot, and mocking patterns. Use when creating tests, writing specs for models/controllers/requests, setting up test fixtures, or mocking external services.
Audio Transcriber
yousufjoyian
Claude agent skills and workspace configuration
change-management-plan
mohitagw15856
"Create a structured change management plan for any organisational change. Use when asked to write a change management plan, manage a change initiative, plan a system rollout, or lead an organisational transformation. Produces a plan covering stakeholder analysis, impact assessment, communication strategy, and resistance management."
renaissance-statistical-arbitrage
copyleftdev
Build trading systems in the style of Renaissance Technologies, the most successful quantitative hedge fund in history. Emphasizes statistical arbitrage, signal processing, and rigorous scientific methodology. Use when developing alpha research, signal extraction, or systematic trading strategies.