1,847 free skills
AI & ML skills
Skills for AI/ML work — LLM prompt engineering, RAG pipelines, vector databases, fine-tuning, and agent orchestration.
Sourced from real, public repositories — synced daily, never invented.
1,847 free skills
Skills for AI/ML work — LLM prompt engineering, RAG pipelines, vector databases, fine-tuning, and agent orchestration.
Sourced from real, public repositories — synced daily, never invented.
15 tools across six categories
13 of them never send your data anywhere
Free · No signup · No trial clock
SEE THE DIRECTORY
exploring-llm-evaluations
posthog/posthog
>
exploring-llm-traces
posthog/posthog
>
agent-evaluation
databricks-solutions/lakebase-online-ml
Use this when you need to EVALUATE OR IMPROVE or OPTIMIZE an existing LLM agent's output quality - including improving tool selection accuracy, answer quality, reducing costs, or fixing issues where the agent gives wrong/incomplete responses. Evaluates agents systematically using MLflow evaluation with datasets, scorers, and tracing. IMPORTANT - Always also load the instrumenting-with-mlflow-tracing skill before starting any work. Covers end-to-end evaluation workflow or individual components (tracing setup, dataset creation, scorer definition, evaluation execution).
llm-development
meleantonio/chernycode
LLM and ML development best practices with LangChain and transformers. Use when building AI/ML applications.
skillpkg
miles990/skillpkg
Agent Skills Package Manager - Install once, use everywhere. Manage, share, and sync AI agent skills across platforms.
hyperliquid-net
jkorf/cryptoexchange-skills-hub
Build C#/.NET HyperLiquid integrations with HyperLiquid.Net, including SpotApi, FuturesApi, REST, websocket subscriptions and queries, market data, balances, account ledger, staking, vaults, transfers, withdrawals, builder fees, HIP-3 DEX parameters, spot and perpetual orders, TWAP, triggers, TP/SL, leverage, positions, HyperLiquidCredentials, native symbol formats, dependency injection, HttpResult, WebSocketResult, QueryResult, trackers, and SharedApis. Use for HyperLiquid market data, account, trading, perpetual futures, websocket, error handling, or idiomatic HyperLiquid.Net code.
crabbox
bastani-inc/atomic
Detect and use Crabbox for repository tests and validation on remote runners. Use when crabbox.yaml or .crabbox.yaml exists, the crabbox CLI is available, or work needs remote compute, a clean or reusable environment, target-platform coverage, or auditable execution evidence.
a2a-agents
majiayu000/claude-skill-registry
AI agent and chatbot patterns including personas, pipelines, and messaging integrations. Use when building conversational agents, creating AI personas, designing insight pipelines, or integrating with Slack/Kakao platforms.
3d-cv-labeling-2026
majiayu000/claude-skill-registry
Expert in 3D computer vision labeling tools, workflows, and AI-assisted annotation for LiDAR, point clouds, and sensor fusion. Covers SAM4D/Point-SAM, human-in-the-loop architectures, and vertical-specific training strategies. Activate on '3D labeling', 'point cloud annotation', 'LiDAR labeling', 'SAM 3D', 'SAM4D', 'sensor fusion annotation', '3D bounding box', 'semantic segmentation point cloud'. NOT for 2D image labeling (use clip-aware-embeddings), general ML training (use ml-engineer), video annotation without 3D (use computer-vision-pipeline), or VLM prompt engineering (use prompt-engineer).
_data-ai-mastery
majiayu000/claude-skill-registry
Master Python programming, Data Science, AI/LLM engineering, and Database administration. Use this as a central index to access specialized sub-skills.
a2a-mcp-integration
majiayu000/claude-skill-registry
Integration patterns for combining Agent-to-Agent (A2A) Protocol with Model Context Protocol (MCP) for hybrid agent communication. Use when building systems that need both agent-to-agent communication and agent-to-tool integration, implementing composite architectures, or when user mentions A2A+MCP integration, hybrid protocols, or multi-agent tool access.
crawl4ai
majiayu000/claude-skill-registry
This skill should be used when users need to scrape websites, extract structured data, handle JavaScript-heavy pages, crawl multiple URLs, or build automated web data pipelines. Includes optimized extraction patterns with schema generation for efficient, LLM-free extraction.
Qwen-Ollama
majiayu000/claude-skill-registry
Using Qwen 2.5 models via Ollama for local LLM inference, text analysis, and AI-powered automation
maplibre-expressions
boettiger-lab/agent-skills
Non-obvious MapLibre GL JS expression gotchas: legacy vs modern filter syntax, when to use 'match' vs 'in' for categorical membership, and other traps LLMs commonly fall into. Use when writing filters, paint expressions, or default_filter/default_style config for MapLibre layers.
venice-embeddings
sediman-agent/openskynet
Call POST /embeddings on Venice. Covers request shape (input, model, encoding_format, dimensions, user), OpenAI compatibility, response compression (gzip/br), and practical usage for retrieval, clustering, and RAG.
gemini-embedding-2
jjat00/jjat-skills
Use this skill when generating embeddings with Google's Gemini Embedding 2 model (gemini-embedding-2-preview), building semantic search systems, creating multimodal embeddings (text, images, audio, video, PDF), integrating embeddings with vector databases like Qdrant or ChromaDB, or implementing RAG pipelines with Gemini embeddings. Trigger this skill whenever the user mentions Gemini embeddings, multimodal embeddings, semantic search with Google models, vector search with Gemini, embedding images/audio/video/PDFs, or cross-modal retrieval — even if they don't explicitly say 'embedding'.
netlify-agent-runner
practicalswan/agent-skills
Run AI agent tasks remotely on Netlify using Claude, Codex, or Gemini. Use when the user wants to run an AI agent on their site, get a second opinion from another model, or delegate development tasks to run remotely against their repo.
netlify-ai-gateway
practicalswan/agent-skills
Use OpenAI, Anthropic, Google Gemini, or OpenRouter models from Netlify Functions or Edge Functions without managing provider API keys or accounts — the gateway injects credentials automatically. Reach for this when you add an AI chatbot or completion endpoint, generate images or text with Gemini/GPT/Claude, summarize form submissions with AI, build an LLM-backed API route, stream a long AI generation, or wire up any server-side AI provider call on Netlify. Covers provider SDK setup, injected env vars, model availability, rate limits, credit costs, streaming for long generations, and local dev with netlify dev or the Vite plugin.
firebase-storage
evolv3ai/claude-skills-archive
|
firebase-security
narlyseorg/superhackers
Security assessment methodology for Google Firebase applications, covering Firestore, Realtime Database, Cloud Storage, and Cloud Functions.
choose-storage
khadinakbarlabs/expo-mobile-app-builder
Decide storage layer for an Expo app: expo-sqlite + drizzle vs MMKV vs SecureStore vs AsyncStorage vs Zustand persist. Use when the user says 'storage', 'database', 'where to save data', 'AsyncStorage vs MMKV', 'expo-sqlite'.
stripe-recipes
workato-devs/recipe-skills
Stripe payment integration recipes for Workato. Enables AI agents to generate valid recipe JSON for Stripe operations including customer management, payment processing, and refunds.
stripe-projects
orc-mvp/playorcnroll
Build AI agents and apps that call the Stripe API using Stripe's official AI toolkits (agent-toolkit, ai-sdk, token-meter) and the Stripe MCP server. Use when the user wants to give an LLM/agent the ability to create customers, products, prices, payment links, invoices, or subscriptions through Stripe, or to meter token usage for billing. Sourced from github.com/stripe/ai.
tax-compliance
appeeky/stripe-skills
Audits Stripe Tax configuration, tax collected, nexus/registration coverage, and VAT/GST/sales-tax handling. Use when the user mentions "Stripe Tax", "sales tax", "VAT", "GST", "tax compliance", "nexus", or "am I collecting tax correctly".
mcp-scraping
heldernoid/scrapping
Expose the scraping toolkit as MCP tools so any MCP-compatible AI agent can autonomously discover, configure, and execute web scrapes. Use this to connect Claude Desktop, Claude Code, or any MCP client to the scraping engine.
scrape-leads
nickjwells/skills-workspace
Scrape and verify business leads using Apify, classify with LLM, enrich emails, and save to Google Sheets. Use when user asks to find leads, scrape businesses, generate prospect lists, or build lead databases for any industry or location.
agent-access-checker
jinnworks/jinn-skills
Audit a site's robots.txt, llms.txt, and crawlability against the named AI crawlers — GPTBot, ClaudeBot, Google-Extended, PerplexityBot, and more — for a per-crawler verdict, an llms.txt structural check, and ranked fixes. Use when confirming AI agents can actually reach and read a site. Sharpest when connected to Jinn's Brand DNA over MCP.
threejs-shaders
bullish0x/gamestudio
Three.js shaders - GLSL, ShaderMaterial, uniforms, custom effects. Use when creating custom visual effects, modifying vertices, writing fragment shaders, or extending built-in materials.
indexion-plan-readme
trkbt10/indexion-skills
Generate README documentation writing plans and tasks. Use when the user wants to create README files for packages, plan documentation writing, or generate doc tasks for manual or LLM authoring.
tins
mushroomfleet/tins-skill
creating TINS-compliant README files, generating implementations from TINS READMEs, and validating TINS documentation. Use when users want to build TINS README files from scratch, generate working code from an existing TINS README, validate or improve a TINS README for completeness, or learn the TINS distribution paradigm. Triggers on phrases like "TINS README", "there is no source", "create a README for LLM generation", "generate code from this README", "validate my TINS doc", or when a user wants to describe software so an AI can build it.
regex-vs-llm-structured-text
lloredia/everything-claude-code
Decision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
regex-vs-llm-structured-text
mark393295827/house-maint-ai
Decision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
regex-vs-llm-structured-text
khetansarvesh/ai_skills_repo
Decision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
regex-vs-llm-structured-text
jamkris/everything-gemini-code
Decision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
bilibili-cli
whizzest/bilibili-cli
CLI skill for Bilibili (哔哩哔哩, B站) with token-efficient YAML output for AI agents to browse videos, users, search, trending, dynamics, favorites, and interactions from the terminal
coldbox-ai-integration
coldbox/skills
Use this skill when integrating AI capabilities into a ColdBox application using the BoxLang AI library (bx-ai module) -- including simple chat, streaming, pipelines, agents, RAG with vector memory, document loading, tool calling, and injecting the AI service into handlers or models.
portfolio-monitor
dzianisv/nonfinancialadviceagents
Living DISCIPLINE monitor over an already-reviewed equity book (CSV with theses + written triggers). Pulls live prices via yfinance, computes extension vs 200d MA (euphoria gauge) and % from 52-week high, detects which written triggers in Price_Flag/Next_Step have FIRED or are NEAR (best-effort regex, requires literal $; handles macro proxies BTC/Brent/oil/TON), flags EUPHORIA (>30% above 200dMA + HIGH AI_Bubble_Fragility), UNACTIONED SELL, and CONCENTRATION (>10% of book). Emits stocks/monitor-report-<date>.md: what needs action now, sorted by $. Materially-changed positions are then handed to multi-lens-quorum (separate agent step) for the buy/sell/hold call. Run: python3 scripts/portfolio_monitor.py --csv <file> --out <dir>. Triggers: monitor my portfolio, what changed in my book, did any triggers fire, check my holdings for euphoria or concentration, portfolio discipline check. Discipline not alpha; notification-only; never trades.
particles-physics
tanepiper/teskooano
Physics simulation for particle systems—forces (gravity, wind, drag), attractors/repulsors, velocity fields, turbulence, and collision. Use when particles need realistic or artistic motion, swarm behavior, or field-based animation.
finkg
joker-of-gotham/finkg
构建、扩展或审计中文金融知识图谱时使用。用 LazySearch 检索并把返回全量落盘，深挖成可回溯原文的原子事实，写入 Neo4j。节点按维度组带丰富属性（财务/行情/估值/所有权/供应/治理/风险），边是具体中文金融动作并带机制与量化属性，纵深由机制问题驱动（跨语义层 6~10+ 跳独立路径），并逐单元格核账检索信息的利用率。分阶段与人对齐，无 Gate、签名或密钥。关键词 financial knowledge graph, Neo4j, LazySearch, graph RAG, 产业链, 传导链, 财报, 股东, 多跳推理。
llmwiki
git-fg/llmwiki
|
omg
jeo-tech-ai/oh-my-gods
OMG — Integrated AI agent orchestration skill. Plan with ralph+plannotator, execute with team/bmad, verify browser behavior with agent-browser, apply UI feedback with agentation(annotate), auto-cleanup worktrees after completion. Supports Claude, Codex, Gemini CLI, and OpenCode. Install: ralph, omc, omx, ohmg, bmad, plannotator, agent-browser, agentation.
autoresearch-paper
wdblink/autoresearch-paper
Turn a paragraph-level research brief into a research-first autonomous paper pipeline. Use when the user wants "帮我把这个课题写成论文", "autoresearch 写 paper", "先把算法做出来再写论文", or a multi-hour/multi-day Mavis plan with evaluator freeze, implementation/experiment loop, research acceptance gate, L0/L1/L2 heartbeat, pause/resume/stop, and manifest cleanup.
fp-go-mcp
ibm/fp-go
Use this skill when working with the fp-go MCP (Model Context Protocol) server located in github.com/IBM/fp-go/gen/v2. Trigger on mentions of MCP server, fp-go MCP tools, list_skills, use_skill, search_examples, get_example, configuring fp-go for Claude Desktop, or when the user needs to access fp-go examples and skills programmatically. This skill explains how to configure and use the MCP server to access fp-go documentation, examples, and skills.
prompt-engineer
slantview/claude
Optimizes prompts for LLMs and AI systems. Use when building AI features, improving agent performance, or crafting system prompts. Expert in prompt patterns and techniques.
prompt-engineer
agentic-assets/agent-skills
Use when designing prompts for LLMs, optimizing model performance, building evaluation frameworks, or implementing advanced prompting techniques like chain-of-thought, few-shot learning, or structured outputs.
prompt-engineer
zroman88/claude-skills-to-cursor
Writes, refactors, and evaluates prompts for LLMs — generating optimized prompt templates, structured output schemas, evaluation rubrics, and test suites. Use when designing prompts for new LLM applications, refactoring existing prompts for better accuracy or token efficiency, implementing chain-of-thought or few-shot learning, creating system prompts with personas and guardrails, building JSON/function-calling schemas, or developing prompt evaluation frameworks to measure and improve model performance.
prompt-engineer
jiho00seo-dev/antigravity-awsome-skill
Transforms user prompts into optimized prompts using frameworks (RTF, RISEN, Chain of Thought, RODES, Chain of Density, RACE, RISE, STAR, SOAP, CLEAR, GROW)
prompt-engineer
igunublue/awesome-ai-skills
Expert in designing effective prompts for LLM-powered applications. Masters prompt structure, context management, output formatting, and prompt evaluation. Use when: prompt engineering, system prompt, few-shot, chain of thought, prompt design.
prompt-engineer
promptraft/promptraft-skills
Transforms user prompts into optimized prompts using frameworks (RTF, RISEN, Chain of Thought, RODES, Chain of Density, RACE, RISE, STAR, SOAP, CLEAR, GROW)
llm-app-patterns
karim-bhalwani/agent-skills-collection
Production LLM application patterns, architectures, and best practices. Covers RAG pipelines, agent architectures, prompt engineering, LLMOps, and production deployment patterns.
prompt-engineer
skingford/skills
Prompt engineering methodology for LLM applications. Use when crafting system prompts, designing few-shot examples, building chain-of-thought reasoning, or evaluating prompt quality. Triggers on tasks involving prompt writing, LLM integration, AI application development, or prompt optimization.
prompt-engineer
thienanthx/skill-web-design
Writes, refactors, and evaluates prompts for LLMs — generating optimized prompt templates, structured output schemas, evaluation rubrics, and test suites. Use when designing prompts for new LLM applications, refactoring existing prompts for better accuracy or token efficiency, implementing chain-of-thought or few-shot learning, creating system prompts with personas and guardrails, building JSON/function-calling schemas, or developing prompt evaluation frameworks to measure and improve model performance.
prompt-engineer
yo-steven/claude-skills-exploration-20260522
Writes, refactors, and evaluates prompts for LLMs — generating optimized prompt templates, structured output schemas, evaluation rubrics, and test suites. Use when designing prompts for new LLM applications, refactoring existing prompts for better accuracy or token efficiency, implementing chain-of-thought or few-shot learning, creating system prompts with personas and guardrails, building JSON/function-calling schemas, or developing prompt evaluation frameworks to measure and improve model performance.
prompt-engineer
zacklecon/claude-skills
Use when designing prompts for LLMs, optimizing model performance, building evaluation frameworks, or implementing advanced prompting techniques like chain-of-thought, few-shot learning, or structured outputs.
prompt-engineering
fivents/one-id
Designs and optimizes prompts for Claude models ensuring clarity, structure, and token efficiency. Use when generating or improving prompts for APIs, workflows, or automation.
RAG Pipeline Builder
eddiebelaval/squire
Build retrieval-augmented generation systems that ground LLM responses in your data
dev-rag-pipeline-designer
khalilbenaz/claude-skills-collection
Conception de pipelines RAG (Retrieval-Augmented Generation) — architecture, chunking, embeddings, vector stores, retrieval hybride, re-ranking, évaluation RAGAS. Se déclenche avec "RAG", "retrieval augmented", "vector database", "embeddings", "knowledge base", "Pinecone", "ChromaDB", "Weaviate", "chercher dans mes documents". Also triggers on "RAG pipeline", "chunking and embeddings", "vector search retrieval".
rag-test
quilibrium-community/quily
Test and debug the RAG retrieval pipeline against the live Supabase database. Use when debugging retrieval issues, verifying temporal/livestream query handling, checking what chunks the bot would return for a query, or auditing Discord vs web UI code path consistency. Trigger on phrases like "test the RAG", "what would the bot return for", "debug retrieval", "check if the bot gets the right livestream", "test a query", "rag-test", or any investigation of why the chatbot returned wrong sources.
pinecone
theo-one-ai/hermes-ccc
Managed vector database for production RAG — serverless and pod-based deployment, hybrid search, namespaces, and metadata filtering.
rag-builder
wellux/claude-code-deprecated
>