LLMs, embeddings, RAG, fine-tuning, and AI model tools
48 boosters
"description": "Harness-native ECC plugin for engineering teams - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses", "name": "Affaan Mustafa", "url": "https://x.com/affaanmustafa"
  
  
This skill guides developers in building LLM-powered applications using Claude and Anthropic SDKs, with smart triggering based on API imports and user intent. Ideal for developers looking to integrate Claude into their projects with language-specific best practices.
This skill helps you build LLM-powered applications with Claude. Choose the right surface based on your needs, detect the project language, then read the relevant language-specific documentation. Scan the target file (or, if no target file, the prompt and project) for non-Anthropic provider markers
"name": "claude-mem", "version": "10.4.1", "description": "Persistent memory system for Claude Code - seamlessly preserve context across sessions",
This booster enables Claude Code to process, convert, and extract data from documents (PDF, DOCX, XLSX, PPTX, HTML, images) using the Nutrient DWS API, including OCR, editing, signing, and form-filling capabilities. Developers building document automation workflows benefit from seamless integration with multiple file formats.
"name": "mempalace", "description": "Give your AI a memory — mine projects and conversations into a searchable palace. 19 MCP tools, auto-save hooks, and guided setup.", "name": "milla-jovovich"
"description": "Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.", "name": "Julius Brussee", "url": "https://github.com/JuliusBrussee"
"name": "beads-marketplace", "description": "Local marketplace for beads plugin development", "name": "Steve Yegge"
The most comprehensive open-source library of Claude Code skills and agent plugins — also works with OpenAI Codex, Gemini CLI, Cursor, and 7 more coding agents. Reusable expertise packages covering engineering, DevOps, marketing, compliance, C-level advisory, and more.  tool for downloading, uploading, and managing models, datasets, and Spaces directly from Claude Code. Essential for developers integrating Hugging Face resources into AI workflows.
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.
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
estimates the required memory for inference, including model weights and an optional KV cache, for Safetensors and GGUF for models on the Hugging Face Hub using HTTP Range requests i.e., without downloading or loading any weights locally. Run with pointing to the Hugging Face Hub repository which w
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 :
Search the Hugging Face Hub for llama.cpp-compatible GGUF repos, choose the right quant, and launch the model with or . 1. Search the Hub with . 3. Prefer the exact HF local-app snippet and quant recommendation when it is visible.
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
Finds the best models for a task by querying official HF benchmark leaderboards, enriching results with model size data, filtering for what fits on the user's device, and returning a comparison table with benchmark scores.
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,
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
Run any workload on fully managed Hugging Face infrastructure. No local setup required—jobs run on cloud CPUs, GPUs, or TPUs and can persist results to the Hugging Face Hub. Use this skill when users want to: When assisting with jobs:
"name": "huggingface-skills", "description": "Agent Skills for AI/ML tasks including dataset creation, model training, evaluation, and research paper publishing on Hugging Face Hub", "name": "Hugging Face"
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 :
"name": "understand-anything", "description": "AI-powered codebase understanding — analyze, visualize, and explain any project", "homepage": "https://github.com/Lum1104/Understand-Anything",
"name": "claude-obsidian", "description": "Claude + Obsidian knowledge companion. Sets up a persistent, compounding wiki vault (Karpathy's LLM Wiki pattern). v1.7 \"Compound Vault\" + v1.8 methodology modes close 5 of 5 priority gaps from the May 2026 compass artifact. Ships: substrate alignment wit
You are the ZCF Configuration Architect for the ZCF (Zero-Config Code Flow) project.
You are the ZCF Template Engine Specialist for the ZCF (Zero-Config Code Flow) project.
Strata is an MCP server enabling progressive, scalable tool usage for AI agents across Claude Desktop, Claude Code, and Cursor. It benefits developers building multi-step AI applications that need flexible, composable tool orchestration.
DAOs that grant voting power = current token balance (not snapshot of past balance) are vulnerable: This is what happened to Beanstalk (April 2022, $182M loss) — attacker flash-borrowed Beanstalk gov tokens, voted to drain the treasury, repaid the loan. Same block. ERC20Votes lets users delegate. If
"name": "byterover-cli", "description": "ByteRover's CLI", "cli", "ai", "llm", "mcp", "developer-tools",
A Chinese academic paper editing prompt that removes AI-generated traces and humanizes writing to match natural scholar voice. Benefits researchers, students, and academics writing in Chinese who want their papers to sound authentically human rather than AI-generated.
Yielding Bear provides a single unified API that routes every LLM request to the cheapest capable model across 16+ providers — saving 60-80% vs calling OpenAI, Anthropic, or Google directly. 1. Get an API key at https://yieldingbear.com/api 2. Set environment variable:
"version": "9.18.0", "description": "v9.18.0 — Multi-LLM orchestration for Claude Code with Double Diamond workflows, provider routing, safety gates, and automation. Use '/octo:auto' or just describe what you need. Run /octo:setup.", "repository": "https://github.com/nyldn/claude-octopus",
"name": "memodb-io", "source": "./src/packages/claude-code/plugin", "description": "Skill memory layer for Claude Code - auto-capture, learn, and reuse skills"
TinyAGI supports running multiple AI agents simultaneously, each with its own isolated workspace, configuration, and conversation state. This allows you to have specialized agents for different tasks while maintaining complete isolation. The agent management feature enables you to: When a message ar
This wiki is maintained entirely by your coding agent. No API key or Python scripts needed — just open this repo in Codex, OpenCode, or any agent that reads this file, and talk to it. Describe what you want in plain English: Or use shorthand triggers:
"$schema": "https://anthropic.com/claude-code/marketplace.schema.json", "description": "Professional multi-agent development workflows with OmO orchestration, Requirements-Driven and BMAD methodologies", "email": "evanxian9@gmail.com"
Diagnose and fix the most common Granola issues. Each error includes platform-specific symptoms, root causes, and step-by-step remediation. Granola captures audio from your device's system audio output (not via meeting platform APIs), so most issues trace back to audio permissions or device configur
Detect and analyze arbitrage opportunities across cryptocurrency exchanges and DeFi protocols. Aggregates prices from CEX and DEX sources, calculates net profit after fees, and identifies direct, triangular, and cross-chain arbitrage paths. 1. Quick spread scan on a specific pair: Shows current pric
Employee onboarding workflow: create employee records, manage documents, handle country-specific compliance requirements.
Witsy is a cross-platform Electron-based desktop AI assistant that serves as a universal MCP (Model Context Protocol) client. Built with Electron, TypeScript, Vue 3, and Vite, it integrates multiple LLM providers and supports features like chat completion, image generation, speech-to-text, text-to-s
This skill provides automated assistance for react context setup tasks within the Frontend Development domain. This skill activates automatically when you: 1. Provides step-by-step guidance for react context setup
"name": "taches-cc-resources", "description": "Curated Claude Code skills and commands for prompt engineering, MCP servers, subagents, hooks, and productivity workflows", "name": "Lex Christopherson",
End-to-end migration playbook for moving from SAP Concur or legacy travel management systems to Navan. Navan uses REST APIs with OAuth 2.0 — there is no SDK, no automated migration tool, and no sandbox for testing. 1. Configure SCIM connector in your IdP (Okta, Azure AD) 2. Map IdP groups to Navan r
<!-- AUTO-GENERATED FILE. DO NOT EDIT MANUALLY. --> <!-- Source: tools/ai/llms-base.txt + tools/ai/aicontextheader.md --> <!-- Regenerate with: python tools/ai/generateaicontext_files.py -->
Unity-MCP generates custom requirement checklists that validate the quality and completeness of feature specifications rather than testing implementation. Developers and product teams benefit by ensuring requirements are clear, specific, and testable before development begins.
"description": "Cloud-backed persistent memory powered by Deeplake — read, write, and share memory across Claude Code sessions and agents", "version": "0.7.104", "name": "Activeloop",
