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
docstring-coverage
siddham-jain/docstring-coverage
>
sqlite-optimization
jrajasekera/jr-agent-skills
Optimize SQLite database performance through configuration, schema design, indexing, and query tuning. Use when users ask to improve SQLite speed, reduce latency, optimize queries, configure PRAGMAs, fix slow queries, handle concurrency, optimize writes/inserts, or tune SQLite for production. Triggers on mentions of SQLite performance, slow queries, PRAGMA settings, WAL mode, indexing strategies, bulk inserts, or database maintenance (VACUUM, ANALYZE).
nocobase-ai-knowledge-base-manager
nocobase/skills
Use when users need to check Professional+ knowledge-base capability, consent to enabling an installed-disabled KB plugin, or manage NocoBase vector databases, Local/Readonly/External knowledge bases, documents, retrieval tests, and binding preparation through nb api kb.
hyde-retrieval
yonatangross/orchestkit
HyDE (Hypothetical Document Embeddings) for improved semantic retrieval. Use when queries don't match document vocabulary, retrieval quality is poor, or implementing advanced RAG patterns.
defense-in-depth
yonatangross/orchestkit
Use when building secure AI pipelines or hardening LLM integrations. Defense-in-depth implements 8 validation layers from edge to storage with no single point of failure.
context-engineering
yonatangross/orchestkit
Use when designing agent system prompts, optimizing RAG retrieval, or when context is too expensive or slow. Reduces tokens while maintaining quality through strategic positioning and attention-aware design.
bmad-advanced-elicitation
bmad-labs/skills
Push the LLM to reconsider, refine, and improve its recent output. Use when user asks for deeper critique or mentions a known deeper critique method, e.g. socratic, first principles, pre-mortem, red team.
tensorrt-llm
nousresearch/hermes-agent
High-throughput LLM inference on NVIDIA GPUs.
instructor
nousresearch/hermes-agent
Structured LLM outputs validated with Pydantic.
nemo-curator
nousresearch/hermes-agent
Curate LLM training data: dedupe, filter, PII redaction.
LangChain Structured Output & HITL
langchain-ai/langchain-skills
INVOKE THIS SKILL when you need structured/typed output from LLMs OR human-in-the-loop approval. Covers with_structured_output(), Pydantic schemas, union types for multiple formats, and HITL middleware. CRITICAL: Fixes for accessing structured response wrong, missing field descriptions, and Pydantic v1 vs v2.
llm-streaming
yonatangross/orchestkit
LLM streaming response patterns. Use when implementing real-time token streaming, Server-Sent Events for AI responses, or streaming with tool calls.
bmad-create-architecture
bmad-labs/skills
Create architecture solution design decisions for AI agent consistency. Use when the user says "lets create architecture" or "create technical architecture" or "create a solution design"
boxlang-runtime-cli-scripting
ortus-boxlang/skills
Use this skill when writing BoxLang CLI scripts and classes, handling command-line arguments, using the BoxLang REPL, running action commands (compile, cftranspile, featureaudit), detecting runtime context, and leveraging CLI-specific built-in functions.
firebase-ai-logic
firebase/firebase-tools
Integrate Firebase AI Logic (Gemini in Firebase) for intelligent app features. Use when adding AI capabilities to Firebase apps, implementing generative AI features, or setting up Firebase AI SDK. Handles Firebase AI SDK setup, prompt engineering, and AI-powered features.
bmad-distillator
bmad-labs/skills
Lossless LLM-optimized compression of source documents. Use when the user requests to 'distill documents' or 'create a distillate'.
workflow
alicoder001/agent-skills
AI agent operational rules including token discipline, navigation-first approach, and output contracts. Use when you need efficient and predictable agent behavior during development tasks.
genkit
akillness/oh-my-gods
Build production-ready AI workflows using Firebase Genkit. Use when creating flows, tool-calling agents, RAG pipelines, multi-agent systems, or deploying AI to Firebase/Cloud Run. Supports TypeScript, Go, and Python with Gemini, OpenAI, Anthropic, Ollama, and Vertex AI plugins.
review-soul-sp-bp
binbingoes/review-soul-sp-bp
Review and self-check smart-hardware and AI-hardware company, business-line, product-line, platform, or functional SP/BP plans against an approved SOUL template plus goal.md, adaptive operating baselines, SWOT/TOWS, Ansoff matrix, growth curves, Where to Play/How to Win, E0-E4 evidence maturity, hardware economics, trust, AI Native leverage, and staged gates. Use when executives, circle leads, product owners, finance teams, or agents ask to 检查SP、检查BP、SPBP自检、智能硬件战略评审、AI硬件规划、硬件产品线审阅、跨圈组合审阅、预算评审、经营指标基线、财务基线、SP牵引指标、业务规划体检、产品线规划复盘、SOUL原则检查, or need a consistent human/agent verdict of 通过、验证、暂停、终止.
prompt-engineer
solatis/claude-config
Invoke IMMEDIATELY via python script when user requests prompt optimization. Do NOT analyze first - invoke this skill immediately.
peft-fine-tuning
nousresearch/hermes-agent
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.
fine-tuning-with-trl
nousresearch/hermes-agent
TRL: SFT, DPO, PPO, GRPO, reward modeling for LLM RLHF.
databricks-ai-functions
databricks-solutions/ai-dev-kit
Use Databricks built-in AI Functions (ai_classify, ai_extract, ai_summarize, ai_mask, ai_translate, ai_fix_grammar, ai_gen, ai_analyze_sentiment, ai_similarity, ai_parse_document, ai_prep_search, ai_query, ai_forecast) to add AI capabilities directly to SQL and PySpark pipelines without managing model endpoints. Also covers document parsing and building custom RAG pipelines (parse → prep_search → index → query).
cms-environments-publishing
contentstack/contentstack-agent-skills
Advise developers on configuring environments, publishing content, using delivery and preview tokens, leveraging the Sync API, and understanding CDN and publish queue behavior in Contentstack.
managing-path-cleaning-rules
posthog/skills
Inspects URL paths and proposes, tests, orders, and applies project-level path cleaning rules so dynamic segments (numeric IDs, UUIDs, slugs, dates) collapse into readable aliases. Use when the user says "clean the paths", "normalize URLs", "group similar pages", "too many distinct paths", "/users/123 and /users/456 are the same page", "set up path cleaning", or asks why a Web analytics or Paths breakdown is fragmented across thousands of nearly-identical URLs. Covers regex syntax (re2), alias placeholder convention, rule ordering, the test workflow, and applying rules via the project-settings-update MCP tool.
ideate
athola/claude-night-market
Generate diverse solution candidates with category-spanning ideation methods and rotation. Use when stuck on a design or fighting repetitive LLM output.
xaf-office
kashiash/xaf-skills
XAF Office/Document Management Modules - FileAttachmentsModule with IFileData/FileAttachment patterns (XPO and EF Core), SpreadsheetModule for Excel editing with ISpreadsheetValueStorage, RichTextModule for Word-like editing with IRichTextDocumentProvider and mail merge, PdfViewerModule for PDF display, platform differences (Blazor vs WinForms), programmatic document manipulation. Use when adding file attachments, spreadsheet editing, rich text editing, or PDF viewing to DevExpress XAF applications.
ux-interaction-review
nickcrew/claude-cortex
Combined UX and interaction design review: usability heuristics, state coverage, feedback patterns, timing, keyboard behavior, error recovery, and progressive disclosure. Use when reviewing how an interface works mechanically and whether interactions are well-designed for use.
project-init
alicoder001/agent-skills
Explicit setup and context-recovery helper for AI agents. Use when the user asks for onboarding, full setup, or deliberate regeneration of .agents/CONTEXT.md. Keep this as a secondary helper; Codex startup self-heal and automatic AGENTS.md or .agents/CONTEXT.md repair belong to codex-precision.
planning
alicoder001/agent-skills
Task decomposition, goal-oriented planning, and adaptive execution strategies for AI agents. Use when facing complex multi-step tasks that require structured approach.
yaml-jazz
simhacker/moollm
YAML is sheet music. The LLM is the jazz musician. Comments are soul.
CTF•AI/ML 攻防
asdfgh1445/ctf-super-hub
用于对抗样本、模型提取、提示注入、成员推断、训练投毒、LoRA 滥用、LLM 越狱等 AI/ML 相关 CTF 题;触发名:ctf-ai-ml
engineering-ai-pipelines
legout/data-agent-skills
AI/ML production workflows: embedding generation, vector storage, RAG patterns, LLM monitoring, and batch inference.
AgentDB Vector Search
natea/fitfinder
Implement semantic vector search with AgentDB for intelligent document retrieval, similarity matching, and context-aware querying. Use when building RAG systems, semantic search engines, or intelligent knowledge bases.
protocol
simhacker/moollm
K-lines as semantic activation — the name activates the tradition.
rag-pipeline
param087/agent-ml-skills
Use when building retrieval-augmented generation. Covers chunking strategy, embedding choice, vector stores, hybrid + reranking retrieval, prompt assembly, and evaluating retrieval and answer quality.
distributed-llm-pretraining-torchtitan
nousresearch/hermes-agent
Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.
simpo-training
nousresearch/hermes-agent
Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.
qdrant-vector-search
nousresearch/hermes-agent
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.
slime-rl-training
nousresearch/hermes-agent
Provides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM models, implementing custom data generation workflows, or needing tight Megatron-LM integration for RL scaling.
running-claude-code-via-litellm-copilot
xixu-me/skills
Use when routing Claude Code through a local LiteLLM proxy to GitHub Copilot, reducing direct Anthropic spend, configuring ANTHROPIC_BASE_URL or ANTHROPIC_MODEL overrides, or troubleshooting Copilot proxy setup failures such as model-not-found, no localhost traffic, or GitHub 401/403 auth errors.
tools
alicoder001/agent-skills
Dynamic tool selection, composition, and error handling patterns for AI agents. Use when you need to efficiently leverage available tools and handle failures gracefully.
crawl4ai-pipeline-builder
prorise-cool/prorise-claude-skills
当需要使用 Crawl4AI 进行多 URL 抓取、Markdown 生成、结构化抽取、CLI 批量 crawl、会话复用、内容过滤或无 LLM 的 schema 提取流水线时使用。适用于“批量抓多个页面”“把文档站转成 markdown”“用 CSS schema 抽结构化数据”“做可复用的 crawl pipeline”等场景。
llms-txt-and-crawler-access
wakqasahmed/skills
Review crawler and AI-agent access files for ecommerce sites, including robots.txt, sitemap, llms.txt, and AI bot rules.
healthcheck
openclaw/skills
Track water and sleep with JSON file storage
llm-finetuning
param087/agent-ml-skills
Use when fine-tuning a large language model. Covers choosing full vs LoRA/QLoRA, dataset formatting, the transformers/PEFT/TRL stack, key hyperparameters, and evaluating fine-tunes without overfitting.
nemo-speech-asr-finetune
nvidia-nemo/nemo
Guide NeMo Speech users through ASR fine-tuning with container setup and Lhotse training.
mcp-expert
fabriciofs/mcp-postgres
Expert in Model Context Protocol (MCP) server development. Use when building MCP servers, creating tools for Claude, implementing resources, debugging MCP connections, or integrating databases with Claude Code.
lbo-model
nousresearch/hermes-agent
Build leveraged buyout workbooks with IRR/MOIC in Excel.
web-search
inference-sh/skills
Web search and content extraction with Tavily and Exa via inference.sh CLI. Apps: Tavily Search, Tavily Extract, Exa Search, Exa Answer, Exa Extract. Capabilities: AI-powered search, content extraction, direct answers, research. Use for: research, RAG pipelines, fact-checking, content aggregation, agents. Triggers: web search, tavily, exa, search api, content extraction, research, internet search, ai search, search assistant, web scraping, rag, perplexity alternative
wtf
citypaul/.dotfiles
Re-explain the immediately previous LLM response when it did not land, using plain, precise UK English.
ronacher-pragmatic-design
copyleftdev/sk1llz
Write Python code in the style of Armin Ronacher, creator of Flask and Jinja2. Emphasizes pragmatic minimalism, explicit over implicit, and composable design. Use when building frameworks, libraries, or applications that need to be extensible and maintainable.
bilibili-cli
hwj123hwj/custom-skills
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
prompt-engineer
megastep/codex-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.
together-fine-tuning
zainhas/skills_draft
LoRA, full fine-tuning, DPO preference tuning, VLM training, function-calling tuning, reasoning tuning, and BYOM uploads on Together AI. Reach for it whenever the user wants to adapt a model on custom data rather than only run inference, evaluate outputs, or host an existing model.
nebius-datalab-pipeline
arindam200/nebius-skills
Full DataLab end-to-end pipeline on Nebius Token Factory — from raw inference logs through synthetic data generation, fine-tuning, and model deployment. Use this skill whenever the user wants to run a complete MLOps workflow on Nebius, combine DataLab with fine-tuning, do teacher-student distillation at scale, build a data flywheel, automate the path from prompts to a deployed custom model, or orchestrate multiple Nebius services together. Trigger for phrases like "full Nebius pipeline", "DataLab workflow", "end-to-end fine-tuning on Nebius", "teacher distillation pipeline", "data flywheel", "automate Nebius training and deployment", or any question involving multiple Nebius services chained together.
planning-with-files
89jobrien/steve
Transforms workflow to use Manus-style persistent markdown files for planning, progress tracking, and knowledge storage. Use when starting complex tasks, multi-step projects, research tasks, or when the user mentions planning, organizing work, tracking progress, or wants structured output.
brand-icons
sugarforever/boring-video-studio
给视频封面 / 合成 / 幻灯片取官方的 AI、LLM、公司品牌矢量 logo（OpenAI、Anthropic、Gemini、DeepSeek、GLM / 智谱、Qwen、Codex……），从 LobeHub Icons 的静态 SVG CDN 直接拉，不手画也不 AI 生成；提供按关键词搜图标 + 下载到项目的脚本。当用户说「加个 X 的 logo / 品牌图标」「封面放 OpenAI 标」「找 GLM 的 icon」「需要某模型的官方标识」时用本 skill。
add-ai-integration
getsentry/sentry-javascript
Add a new AI provider integration to the Sentry JavaScript SDK. Use when contributing a new AI instrumentation (OpenAI, Anthropic, Vercel AI, LangChain, etc.) or modifying an existing one.
llm-obs-trace-rca
datadog-labs/agent-skills
Root cause analysis on production LLM traces. Diagnoses why an LLM application is failing — works from eval judge verdicts, runtime errors, or structural anomalies depending on what signals are present. Walks the span tree from symptom to root cause. Use when user says "what's wrong with my app", "why is my eval failing", "analyze errors", "root cause analysis", "diagnose failures", or wants to understand production failure patterns.