1.3K skills · sorted by relevance
Guide development of a LangChain or LangGraph project created with agents-cli. It explains tool additions, local runs, evaluation, and serving while identifying the required graph structure and commands that differ from ADK projects.
Audit or revise formulaic prose while preserving facts, voice, and document structure. A local Python script measures selected writing patterns and checks explicit source details, while the assistant reviews context and makes edits.
Revise text to remove formulaic AI phrasing while keeping the writer’s voice, facts, intent, and structure. It targets justified changes, preserves protected passages, and reports editing limits and any unresolved issues.
Update existing Lark/Feishu documents while preserving comments, highlights, citations, and newer remote edits. Maintain a private local copy, apply targeted changes, and verify the updated document for ongoing project use; LoopX integration is optional.
Deploy ML models on Kubernetes with KServe (formerly KFServing) and NVIDIA Triton Inference Server. Includes canary deployments, autoscaling, model versioning, A/B testing, and GPU resource management for production model serving.
The English summary is being prepared. Showing the original description.
LLM and ML model deployment for inference. Use when serving models in production, building AI APIs, or optimizing inference. Covers vLLM (LLM serving), TensorRT-LLM (GPU optimization), Ollama (local), BentoML (ML deployment), Triton (multi-model), LangChain (orchestration), LlamaIndex (RAG), and streaming patterns.
The English summary is being prepared. Showing the original description.
Use when running throughput/latency (serving speed) benchmarks against an already-running model server. Uses sgl_jax.bench_serving to sweep batch-size/concurrency and measure throughput and latency, producing summary.csv and per-point bs_<N>/result.jsonl. Speed only — accuracy/eval belongs to model-accuracy-benchmark. Triggers on throughput, latency, bench_serving, batch-size sweep, tok/s, TTFT, TPOT — not accuracy/eval.
The English summary is being prepared. Showing the original description.
Databricks Model Serving endpoint lifecycle and ops. Use when asked to: CRUD serving endpoints (CLI or MLflow Deployments client); configure traffic routing for A/B / canary deploys and zero-downtime version swaps; retrieve OpenAPI schemas; inspect logs, metrics, or permissions; manage legacy AI Gateway rate limits (not Unity Gateway); discover Foundation Model API endpoints at runtime; integrate endpoints into Databricks Apps; or stream from off-platform clients (Vercel AI SDK v6, standalone Node.js). NOT for: Unity Gateway CRUD and management (databricks-unity-gateway), training, MLflow autologging, UC registration, custom PyFunc/ResponsesAgent authoring (databricks-ml-training); Knowledge Assistants/Supervisor Agents (databricks-agent-bricks); MLflow evaluation (databricks-mlflow-evaluation).
The English summary is being prepared. Showing the original description.
Design privacy-preserving analytics systems using differential privacy, k-anonymity, l-diversity, and t-closeness. Covers privacy budget allocation with epsilon tracking, references Google DP library, OpenDP, and Apple PPML. Includes Python differential privacy implementation for GDPR-compliant statistical analysis.
The English summary is being prepared. Showing the original description.
Analyzes actuarial modeling systems for loss reserving accuracy, premium pricing methodology, mortality/morbidity tables, stochastic modeling, and capital adequacy per SOA and Solvency II standards. USE THIS SKILL WHEN: - You need to review or audit actuarial models (reserving, pricing, capital) - Someone asks about loss triangle analysis or reserve adequacy - You are evaluating IBNR calculations, chain ladder methods, or Bornhuetter-Ferguson - A project involves insurance pricing, GLM rating models, or ratemaking - You need to assess Solvency II SCR calculations or RBC compliance - Someone mentions actuarial opinions, ASOP compliance, or SOA standards - You are reviewing stochastic models, ESG configurations, or DFA frameworks - A codebase uses actuarial libraries (chainladder, lifetables, ChainLadder R package) TRIGGER PHRASES: "actuarial", "loss reserving", "IBNR", "chain ladder", "premium pricing", "mortality table", "Solvency II", "capital adequacy", "ratemaking", "GLM pricing", "risk-based capital", "reserve analysis", "actuarial opinion"
The English summary is being prepared. Showing the original description.
Design and configure model serving infrastructure — endpoint scaling, batching, GPU allocation. Use when asked to "serve this model", "design an inference endpoint", or "size GPU allocation for serving".
The English summary is being prepared. Showing the original description.
Audit current model deployment topology — serving config, latency profile, version inventory. Use when asked "what models are deployed", "audit our serving setup", or "inventory our model versions".
The English summary is being prepared. Showing the original description.
Practical naked-eye and binocular sky observing. Covers dark adaptation, limiting magnitude, constellation recognition, star hopping, the messier and Caldwell catalogs accessible without a telescope, the Moon, planets, meteor showers, aurorae, and the ethics and habits of observing under light-polluted and dark skies. Use when teaching someone to find their way around the sky, planning a naked-eye session, or choosing a first binocular tour.
The English summary is being prepared. Showing the original description.
Deploy machine learning models to production serving infrastructure using MLflow, BentoML, or Seldon Core with REST/gRPC endpoints, implement autoscaling, monitoring, and A/B testing capabilities for high-performance model inference at scale. Use when deploying trained models for real-time inference, setting up REST or gRPC prediction APIs, implementing autoscaling for variable load, running A/B tests between model versions, or migrating from batch to real-time inference.
The English summary is being prepared. Showing the original description.
Build a feature store using Feast for centralized feature management, configure offline and online stores for batch and real-time serving, define feature views with transformations, and implement point-in-time correct joins for ML pipelines. Use when managing features for multiple ML models, ensuring training-serving consistency, serving low-latency features for real-time inference, reusing feature definitions across projects, or building a feature catalog for discovery and governance.
The English summary is being prepared. Showing the original description.
Design model serving infrastructure for AI. TRIGGERS - Use when user needs help with ai-model-serving related tasks.
The English summary is being prepared. Showing the original description.
Pick the right serving container for a SageMaker model deployment and find its current image URI. Use this skill whenever about to deploy a model to a SageMaker endpoint and an image URI needs to be chosen — including when the user says "deploy this LLM", "host this HuggingFace model", "serve this fine-tuned model", "deploy this embedding model", "host a reranker", "serve a sentence-transformers model", or when about to hardcode any container URI in deployment code. HuggingFace-curated Deep Learning Containers are ALWAYS preferred: HuggingFace vLLM (LLMs and generative rerankers), HuggingFace vLLM-Omni (multimodal), TEI (embeddings/cross-encoder rerankers), HF Inference Toolkit (other transformers). Generic images (AWS vLLM, DJL-LMI, SGLang) are used only when no HuggingFace image is compatible — never merely because they carry a newer version. Never hardcode a container URI from memory and never default to TGI. Prevents stale-image failures and wrong-region URIs.
The English summary is being prepared. Showing the original description.
Design model serving infrastructure for AI. TRIGGERS - Use when user needs help with ai-model-serving related tasks.
The English summary is being prepared. Showing the original description.
Measure and improve OpenClaw test and plugin-suite performance by investigating slow imports, repeated startup work, and memory use. Compare results before and after changes while preserving test coverage and verifying the affected tests.
Investigate performance bottlenecks in REA analysis workflows and improve caching, queries, or shared processing. Compare completed tasks before and after changes while preserving correct results, examined coverage, and access to supporting evidence.
Coordinate code refactoring, restructuring code while preserving its behavior. Check existing tests before changes, work in small steps, review the result, and pause for approval of the plan and before committing.
Set up and optimize large language model inference, or response generation, on NVIDIA GPUs. Use quantization, batch processing, and multiple GPUs to improve serving performance, and expose models through a chat completion API.
Plan and implement migrations for data, APIs, configuration, and dependencies while preserving existing data and compatibility. Define rollback steps, check old and new behavior, and handle retries and partial failures during staged transitions.
Review a screen, user flow, or component against the user's task. Assess visual hierarchy, interaction cues, error handling, accessibility, and consistency, then provide prioritized fixes with reasons and identify strengths worth preserving.