490 boosters for "form" — open source, verified from GitHub, ready to install
You are an expert at using the TRL (Transformers Reinforcement Learning) library to train and fine-tune large language models. TRL provides CLI commands for post-training foundation models using state-of-the-art techniques: TRL is built on top of Hugging Face Transformers and Accelerate, providing s
Tiebreakers when the request is ambiguous: "embedding model" / "vector search" / "similarity" → [SentenceTransformer]. "rerank" / "ranker" / "two-stage" → [CrossEncoder]. "SPLADE" / "sparse" / "inverted index" → [SparseEncoder]. If still unclear, ask. Override only if the user specifies otherwise: T
This skill is for running evaluations against models on the Hugging Face Hub on local hardware. It does not cover: If the user wants to run the same eval remotely on Hugging Face Jobs, hand off to the skill and pass it one of the local scripts in this skill.
Train object detection, image classification, and SAM/SAM2 segmentation models on managed cloud GPUs. No local GPU setup required—results are automatically saved to the Hugging Face Hub. Use this skill when users want to: Helper scripts use PEP 723 inline dependencies. Run them with :
Train language models using TRL (Transformer Reinforcement Learning) on fully managed Hugging Face infrastructure. No local GPU setup required—models train on cloud GPUs and results are automatically saved to the Hugging Face Hub. Use this skill when users want to: Use Unsloth () instead of standard
Transformers.js enables running state-of-the-art machine learning models directly in JavaScript, both in browsers and Node.js environments, with no server required. Use this skill when you need to: The pipeline API is the easiest way to use models. It groups together preprocessing, model inference,
Fast, calibration-free weight quantization supporting 8/4/3/2/1-bit precision with multiple optimized backends. HQQ uses to define quantization parameters: The core quantized layer that replaces :
A system prompt for Claude Code that enforces defensive security practices and provides CLI guidance, designed to help developers safely use Claude for software engineering tasks while preventing misuse.
A skill for fine-tuning and training language models on Hugging Face's cloud GPU infrastructure using TRL, supporting SFT, DPO, GRPO methods and GGUF conversion for local deployment. Developers and ML engineers working with cloud-based model training benefit from this comprehensive guidance.
This skill automates the process of adding, extracting, and managing evaluation results in Hugging Face model cards, supporting multiple data sources including Artificial Analysis API and custom evaluations with vLLM/lighteval. It's valuable for ML practitioners and model maintainers who need to track and display model performance metrics.
This skill enables AI assistants to create, configure, and manage datasets on Hugging Face Hub with SQL-based querying and transformation capabilities. It's valuable for developers building data workflows and ML projects that require programmatic dataset management.
1. Fetch homepage HTML (curl or WebFetch) 2. Detect business type (SaaS, Local, E-commerce, Publisher, Agency, Other) 3. Extract key pages from sitemap.xml or internal links (up to 50 pages)
Generates per-category catalog JSON files from the npm package's public API. Each catalog lists all UseCase and EventHandler abstractions in a category with their resolved source file paths. LLMs read source files on demand for exact, up-to-date types — no enrichment phase needed. 1. Discovers all
ByteRover CLI () - Interactive REPL with React/Ink TUI oclif v4, TypeScript (ES2022, Node16 modules, strict), React/Ink (TUI), Zustand, axios, socket.io, Mocha + Chai + Sinon + Nock
❶ One is a new React framework - target web and native with a single Vite plugin and fully shared code, so you can ship cross-platform nearly as easy as single-platform.
This rule describes the Continuous Integration and Deployment (CI/CD) workflow for the Imageflow project, primarily defined in and utilizing helper scripts within the directory.
Your Friendly Open-Source Gen-AI Platform
A physics tutoring booster that adapts explanations from intuitive reasoning to formal derivations based on learner level. Ideal for students, educators, and researchers seeking physics help across all difficulty tiers.
以下是你所需要生成测试用例的对象的描述，也即来自远程MCP服务器的工具描述。你可以使用调用以下工具。 请首先尽可能全面覆盖并输出所有当前威胁的测试维度，而后为测试目标的每个维度设计测试，对于每个维度至少生成3个测试用例。
You are a specialized reviewer for refactoring and code transformations. Your job is to verify that refactoring preserves behavior, improves quality, and follows safe transformation practices. You render verdicts on refactoring quality. Before reviewing, frame the question space E(X,Q): Your task pr
See CLAUDE.md for project conventions and development guidelines.
You are an elite CLAUDE.md auditor and documentation integrity specialist. Your sole purpose is to ensure every file and file in the project accurately reflects the current codebase state. You work autonomously: discover, analyze, and fix documentation drift without manual guidance. You are method
You are an elite code quality engineer specializing in automated code quality remediation. Your sole purpose is to ensure the codebase passes all quality checks by iteratively fixing issues until the codebase is clean. You operate in a loop where each iteration is delegated to a subagent via the Tas
MIRIX is a Multi-Agent Personal Assistant with an Advanced Memory System. It features a six-agent memory architecture (Core, Episodic, Semantic, Procedural, Resource, Knowledge Vault) with screen activity tracking and privacy-first design. 1. Follow PEP 8 strictly 3. Documentation
