Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform

Provides guidance for automatically evolving and optimizing AI agents across any domain using LLM-driven evolution algorithms. Use when building self-improving agents, optimizing agent prompts and skills against benchmarks, or implementing automated agent evaluation loops.

Multi-agent orchestration framework for autonomous AI collaboration. Use when building teams of specialized agents working together on complex tasks, when you need role-based agent collaboration with memory, or for production workflows requiring sequential/hierarchical execution. Built without LangChain dependencies for lean, fast execution.

Managed vector database for production AI applications. Fully managed, auto-scaling, with hybrid search (dense + sparse), metadata filtering, and namespaces. Low latency (<100ms p95). Use for production RAG, recommendation systems, or semantic search at scale. Best for serverless, managed infrastructure.

High-performance vector similarity search engine for RAG and semantic search. Use when building production RAG systems requiring fast nearest neighbor search, hybrid search with filtering, or scalable vector storage with Rust-powered performance.

Framework for state-of-the-art sentence, text, and image embeddings. Provides 5000+ pre-trained models for semantic similarity, clustering, and retrieval. Supports multilingual, domain-specific, and multimodal models. Use for generating embeddings for RAG, semantic search, or similarity tasks. Best for production embedding generation.

Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers, vLLM), and maximize inference speed with Outlines - dottxt.ai's structured generation library

Open-source AI observability platform for LLM tracing, evaluation, and monitoring. Use when debugging LLM applications with detailed traces, running evaluations on datasets, or monitoring production AI systems with real-time insights.

Evaluates NVIDIA Cosmos Policy on LIBERO and RoboCasa simulation environments. Use when setting up cosmos-policy for robot manipulation evaluation, running headless GPU evaluations with EGL rendering, or profiling inference latency on cluster or local GPU machines.

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.

Fine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies for robot action generation with continuous action heads, LoRA adaptation, and FiLM conditioning on LIBERO simulation and ALOHA real-world setups. Use when reproducing OpenVLA-OFT paper results, training custom VLA action heads (L1 or diffusion), deploying server-client inference for ALOHA, or debugging normalization, LoRA merge, and cross-GPU issues.

Foundation model for image segmentation with zero-shot transfer. Use when you need to segment any object in images using points, boxes, or masks as prompts, or automatically generate all object masks in an image.

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.

OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.

Merge multiple fine-tuned models using mergekit to combine capabilities without retraining. Use when creating specialized models by blending domain-specific expertise (math + coding + chat), improving performance beyond single models, or experimenting rapidly with model variants. Covers SLERP, TIES-Merging, DARE, Task Arithmetic, linear merging, and production deployment strategies.

Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT. Use when compressing models without retraining, achieving 50% sparsity with minimal accuracy loss, or enabling faster inference on hardware accelerators. Covers unstructured pruning, structured pruning, N:M sparsity, magnitude pruning, and one-shot methods.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with limited compute. Covers draft models, tree-based attention, Jacobi iteration, parallel token generation, and production deployment strategies.

Generates publication-quality figures for ML papers from research context. Given a paper section or description, extracts system components and relationships to generate architecture diagrams via Gemini. Given experiment results or data, auto-selects chart type and generates data-driven figures via matplotlib/seaborn. Use when creating any figure for a conference paper.

Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. For systems venues (OSDI, NSDI, ASPLOS, SOSP), use systems-paper-writing instead.

Generates conference presentation slides (Beamer LaTeX PDF and editable PPTX) from a compiled paper with speaker notes and talk script. Use when preparing oral talks, spotlight presentations, or invited talks for ML and systems conferences.

Comprehensive guide for writing systems papers targeting OSDI, SOSP, ASPLOS, NSDI, and EuroSys. Provides paragraph-level structural blueprints, writing patterns, venue-specific checklists, reviewer guidelines, LaTeX templates, and conference deadlines. Use this skill for all systems conference paper writing.

Guides researchers through structured ideation frameworks to discover high-impact research directions. Use when exploring new problem spaces, pivoting between projects, or seeking novel angles on existing work.

Applies cognitive science frameworks for creative thinking to CS and AI research ideation. Use when seeking genuinely novel research directions by leveraging combinatorial creativity, analogical reasoning, constraint manipulation, and other empirically grounded creative strategies.

Base skill for Refly ecosystem: creates, discovers, and runs domain-specific skills bound to workflows. Routes user intent to matching domain skills via symlinks, delegates execution to Refly backend. Use when user asks to: create skills, run workflows, automate multi-step tasks, or manage pipelines. Triggers: refly, skill, workflow, run skill, create skill, automation, pipeline. Requires: @refly/cli installed and authenticated.

refly skill list [options]

UX design methodology and external consultation. Use when creating user flows, wireframes, interaction patterns, or getting UX feedback. Provides structured frameworks for user-centered design.

Expert assistance for next-forge — a production-grade Turborepo template for Next.js SaaS apps. Triggers on questions about next-forge installation, setup, architecture, packages, customization, deployment, and development workflows.

Avue 组件代码生成与配置指南。当用户需要创建、配置或生成基于 Avue (avue-crud, avue-form, avue-tree 等) 的 Vue 页面代码时使用此 skill。覆盖 CRUD 表格、表单、树组件、数据展示等全部 Avue 组件。支持 Composition API 和 Options API 两种代码风格。当用户提到 avue、crud表格、动态表单、JSON配置表单、低代码表单、avue-crud、avue-form 等关键词时，务必使用此 skill。

'Audit and update docs/copilot/ documentation to accurately reflect current VS Code AI capabilities. Use when: competitive analysis reveals gaps, product launches new features, docs use outdated framing, or keyword coverage needs strengthening for discoverability by users and AI agents. Produces a gap analysis plus targeted edits across affected files.'

'Review VS Code release notes for style, structure, and completeness. Use for reviewing Insiders or Stable release notes against writing guidelines. Produces an actionable list of recommendations.'

'Check and optimize MetaDescription frontmatter fields in VS Code documentation. Use when auditing, adding, or improving page descriptions for SEO and discoverability. Apply this when making content changes to markdown articles.'

'Create and manage redirects in VS Code documentation when pages are moved, renamed, or deleted. Use when moving docs pages, renaming files, restructuring content, or when the user asks about redirects.'

Provides best practices and guidance for working with Rush monorepos. Use when the user is working in a Rush-based repository, asks about Rush commands (install, update, build, rebuild), needs help with project selection, dependency management, build caching, subspace configuration, or troubleshooting Rush-specific issues.

> Best practices for building agents and agentic applications with Plano — the AI-native proxy and dataplane. Covers configuration, routing, agent orchestration, filter chains, observability, CLI oper

**Version 1.0.0**

**Version 1.0.0**

Generates beautiful, consistent Preline Theme CSS files. Agent interprets user request, runs build script, delivers complete CSS.

Documentation for the Trellis agent system - specialized AI agents for different development phases.

Use this skill when the user is working with an Aspire distributed application and needs to operate the AppHost or its resources through the Aspire CLI: start, restart, stop, or wait on the app; inspect resources, logs, traces, docs, or health; add integrations; manage secrets or config; publish, deploy, or rerun a named pipeline step; initialize Aspire in an existing app; recover missing `.modules` files in a TypeScript AppHost; discover the right frontend URL for Playwright from Aspire state; expose custom dashboard/resource commands; or understand unfamiliar Aspire AppHost APIs in C# or TypeScript. Use it even if they describe the task in terms of an AppHost, resources, dashboard, existing app bootstrap, missing generated modules, Playwright URL discovery, C# API understanding, or local distributed app workflow without explicitly naming Aspire. Do not use it for non-Aspire .NET apps, container-only repos with no AppHost, or ordinary build and test tasks.

Guides dependency version updates by checking nuget.org for latest versions, triggering the dotnet-migrate-package Azure DevOps pipeline, and monitoring runs. Use this when asked to update external NuGet dependencies.

Guide for writing Aspire deployment end-to-end tests. Use this when asked to create, modify, or debug deployment E2E tests that deploy to Azure.

Updates Docker container image tags used by Aspire hosting integrations. Queries registries for newer tags, uses LLM to determine version-compatible updates, and applies changes. Use this when asked to update container image versions.

Reproduces and fixes flaky or quarantined tests. Tries local reproduction first (fast), then falls back to CI reproduce workflow (reproduce-flaky-tests.yml). Use this when asked to investigate, reproduce, debug, or fix a flaky test, a quarantined test, or an intermittently failing test.

Reviews .NET API surface area PRs for design guideline violations. Analyzes api/*.cs file diffs, applies review rules from .NET Framework Design Guidelines and Aspire conventions, and attributes findings to the developer who introduced each API (via git blame). Use this when asked to review API surface area changes.

Review a GitHub pull request for problems. Use when asked to review a PR, do a code review, check a PR for issues, or review pull request changes. Focuses only on identifying problems — not style nits or praise.

Guide for writing Aspire CLI end-to-end tests using Hex1b terminal automation. Use this when asked to create, modify, or debug CLI E2E tests.

Guide for diagnosing GitHub Actions test failures, extracting failed tests from runs, and creating or updating failing-test issues. Use this when asked to investigate GitHub Actions test failures, download failure logs, create failing-test issues, or debug CI issues.

Guide for writing tests for the Aspire Dashboard. Use this when asked to create, modify, or debug dashboard unit tests or Blazor component tests.

Quarantines or disables flaky/problematic tests using the QuarantineTools utility