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
firecrawl
edxeth/superlight-firecrawl-skill
Scrapes and crawls web pages, converting them to clean markdown or structured JSON for LLM consumption. Use when needing to extract content from URLs, crawl entire websites, map site structure, search the web with scraping, or extract structured data from pages. Best for web scraping, site crawling, URL discovery, and converting web content to LLM-ready formats.
mobile-insecure-storage
securityfortech/hacking-skills
>
llm-ollama-integration
findinfinitelabs/chuuk
Ollama LLM integration for Chuukese translation and text generation. Use when working with local LLM models, creating custom modelfiles, or implementing hybrid translation approaches combining Helsinki-NLP with LLMs.
regex-vs-llm-structured-text
angelmic/everything-claude-code-ios
Decision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
ralph-loop
sundial-org/awesome-openclaw-skills
Generate copy-paste bash scripts for Ralph Wiggum/AI agent loops (Codex, Claude Code, OpenCode, Goose). Use when asked for a “Ralph loop”, “Ralph Wiggum loop”, or an AI loop to plan/build code via PROMPT.md + AGENTS.md, SPECS, and IMPLEMENTATION_PLAN.md, including PLANNING vs BUILDING modes, backpressure, sandboxing, and completion conditions.
modal-finetuning
jamesrobmccall/modal_skills
>-
ai-workflow-brand
trree/ark-skills
母 skill。研究 AI 时代个人判断力、方法论、系统能力与品牌资产建设。负责定位总控与任务分流。
prompt-engineering
codealive-ai/ceo-ai-os
Universal prompt engineering techniques for any LLM. Use when crafting, optimizing, or reviewing prompts for AI models. Triggers on requests like "improve this prompt", "write a system prompt", "optimize my instructions", "help me prompt engineer", "audit this prompt", "review my prompt", or when building agentic systems that need structured prompts.
vectorize
mohamedabdallah-14/prompt-to-asset
Convert a raster image to SVG. Three paths — Recraft hosted vectorization, vtracer (multi-color polygon), potrace (1-bit). Optimizes with SVGO; validates path count as a quality signal.
prompt-engineer
akiselev/qt-llm-template
Invoke IMMEDIATELY via python script when user requests prompt optimization. Do NOT analyze first - invoke this skill immediately.
prompt-engineering
inference-sh/skills
Master prompt engineering for AI models: LLMs, image generators, video models. Techniques: chain-of-thought, few-shot, system prompts, negative prompts. Models: Claude, GPT-4, Gemini, FLUX, Veo, Stable Diffusion prompting. Use for: better AI outputs, consistent results, complex tasks, optimization. Triggers: prompt engineering, how to prompt, better prompts, prompt tips, prompting guide, llm prompting, image prompt, ai prompting, prompt optimization, prompt template, prompt structure, effective prompts, prompt techniques
llm-models
inference-sh/skills
Access Claude, Gemini, Kimi, GLM and 100+ LLMs via inference.sh CLI using OpenRouter. Models: Claude Opus 4.5, Claude Sonnet 4.5, Claude Haiku 4.5, Gemini 3 Pro, Kimi K2, GLM-4.6, Intellect 3. One API for all models with automatic fallback and cost optimization. Use for: AI assistants, code generation, reasoning, agents, chat, content generation. Triggers: claude api, openrouter, llm api, claude sonnet, claude opus, gemini api, kimi, language model, gpt alternative, anthropic api, ai model api, llm access, chat api, claude alternative, openai alternative
ai-rag-pipeline
inference-sh/skills
Build RAG (Retrieval Augmented Generation) pipelines with web search and LLMs. Tools: Tavily Search, Exa Search, Exa Answer, Claude, GPT-4, Gemini via OpenRouter. Capabilities: research, fact-checking, grounded responses, knowledge retrieval. Use for: AI agents, research assistants, fact-checkers, knowledge bases. Triggers: rag, retrieval augmented generation, grounded ai, search and answer, research agent, fact checking, knowledge retrieval, ai research, search + llm, web grounded, perplexity alternative, ai with sources, citation, research pipeline
python-sdk
inference-sh/skills
Python SDK for inference.sh - run AI apps, build agents, and integrate with all models. Package: inferencesh (pip install inferencesh). Supports sync/async, streaming, file uploads. Build agents with template or ad-hoc patterns, tool builder API, skills, and human approval. Use for: Python integration, AI apps, agent development, RAG pipelines, automation. Triggers: python sdk, inferencesh, pip install, python api, python client, async inference, python agent, tool builder python, programmatic ai, python integration, sdk python
video-prompting-guide
inference-sh/skills
Best practices and techniques for writing effective AI video generation prompts. Covers: Veo, Seedance, Wan, Grok, Kling, Runway, Pika, Sora prompting strategies. Learn: shot types, camera movements, lighting, pacing, style keywords, negative prompts. Use for: improving video quality, getting consistent results, professional video prompts. Triggers: video prompt, how to prompt video, veo prompts, video generation tips, better ai video, video prompt engineering, video prompt guide, video prompt template, ai video tips, video prompt best practices, video prompt examples, cinematography prompts
experiment-analzyer
datadog-labs/agent-skills
Analyze LLM experiment results. Use when user says "analyze experiment", "experiment analysis", "evaluate experiment", "check experiment metrics", or provides an experiment ID for analysis. Requires experiment_id as argument.
blip-2-vision-language
firecrawl/ai-research-skills
Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text retrieval, or multimodal chat with state-of-the-art zero-shot performance.
outlines
firecrawl/ai-research-skills
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
llamaindex
firecrawl/ai-research-skills
Data framework for building LLM applications with RAG. Specializes in document ingestion (300+ connectors), indexing, and querying. Features vector indices, query engines, agents, and multi-modal support. Use for document Q&A, chatbots, knowledge retrieval, or building RAG pipelines. Best for data-centric LLM applications.
gptq
firecrawl/ai-research-skills
Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.
autogpt-agents
firecrawl/ai-research-skills
Autonomous AI agent platform for building and deploying continuous agents. Use when creating visual workflow agents, deploying persistent autonomous agents, or building complex multi-step AI automation systems.
hqq-quantization
firecrawl/ai-research-skills
Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision without needing calibration datasets, for fast quantization workflows, or when deploying with vLLM or HuggingFace Transformers.
crewai-multi-agent
firecrawl/ai-research-skills
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.
axolotl
firecrawl/ai-research-skills
Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support
skypilot-multi-cloud-orchestration
firecrawl/ai-research-skills
Multi-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers.
sentence-transformers
firecrawl/ai-research-skills
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.
slime-rl-training
firecrawl/ai-research-skills
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.
instructor
firecrawl/ai-research-skills
Extract structured data from LLM responses with Pydantic validation, retry failed extractions automatically, parse complex JSON with type safety, and stream partial results with Instructor - battle-tested structured output library
simpo-training
firecrawl/ai-research-skills
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.
peft-fine-tuning
firecrawl/ai-research-skills
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.
awq-quantization
firecrawl/ai-research-skills
Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.
openrlhf-training
firecrawl/ai-research-skills
High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.
evaluating-llms-harness
firecrawl/ai-research-skills
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.
serving-llms-vllm
firecrawl/ai-research-skills
Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.
long-context
firecrawl/ai-research-skills
Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques. Use when processing long documents (32k-128k+ tokens), extending pre-trained models beyond original context limits, or implementing efficient positional encodings. Covers rotary embeddings, attention biases, interpolation methods, and extrapolation strategies for LLMs.
dspy
firecrawl/ai-research-skills
Build complex AI systems with declarative programming, optimize prompts automatically, create modular RAG systems and agents with DSPy - Stanford NLP's framework for systematic LM programming
knowledge-distillation
firecrawl/ai-research-skills
Compress large language models using knowledge distillation from teacher to student models. Use when deploying smaller models with retained performance, transferring GPT-4 capabilities to open-source models, or reducing inference costs. Covers temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.
qdrant-vector-search
firecrawl/ai-research-skills
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.
nemo-curator
firecrawl/ai-research-skills
GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16× faster), quality filtering (30+ heuristics), semantic deduplication, PII redaction, NSFW detection. Scales across GPUs with RAPIDS. Use for preparing high-quality training datasets, cleaning web data, or deduplicating large corpora.
model-pruning
firecrawl/ai-research-skills
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.
llama-factory
firecrawl/ai-research-skills
Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal support
guidance
firecrawl/ai-research-skills
Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and build multi-step workflows with Guidance - Microsoft Research's constrained generation framework
implementing-llms-litgpt
firecrawl/ai-research-skills
Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen, Mistral). Use when need clean model implementations, educational understanding of architectures, or production fine-tuning with LoRA/QLoRA. Single-file implementations, no abstraction layers.
fine-tuning-with-trl
firecrawl/ai-research-skills
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
distributed-llm-pretraining-torchtitan
firecrawl/ai-research-skills
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.
training-llms-megatron
firecrawl/ai-research-skills
Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.
quantizing-models-bitsandbytes
firecrawl/ai-research-skills
Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers. Works with HuggingFace Transformers.
trusty-memory
bobmatnyc/claude-mpm
Persistent memory palace system with hierarchical storage (palace/wing/room/closet/drawer), progressive retrieval (L0-L3), and temporal knowledge graph for cross-session context
cli-vstash
jr2804/prompts
Local document memory with semantic search for AI-assisted workflows. Use when managing project documentation, codebases, or research papers that need persistent memory across sessions. Triggers on: vstash add/search/ask commands, document ingestion, semantic search, RAG pipelines, local knowledge bases, or configuring vstash for personal projects.
32-seo-growth
minhnv0807/ai-business-skills
Khi nguoi dung muon tang traffic tu SEO, AI SEO, GEO/AEO, Google, ChatGPT, Perplexity, schema, structured data, llms.txt, programmatic SEO, local SEO, directory submissions, Product Hunt, G2/Capterra, backlinks, hoac audit website khong len top. Dung cho 'SEO audit', 'AI search visibility', 'duoc AI trich dan', 'schema markup', 'pSEO', 'local SEO', 'submit directory', 'backlink plan'. Khong dung cho setup Google Business Profile lan dau — dung skill 11-thiet-lap-kenh; khong dung cho viet noi dung bai — dung 36-content-brief.
regex-vs-llm-structured-text
tusharkrbarman/skills-for-open-code
Decision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
CTF•AI/ML 攻防
isloverhatwhite/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
llm-engineering
usermeme/skills
Building features and systems on top of LLMs — designing tools agents can actually use, structured outputs, context budgeting, when multi-agent is worth it, reliability around model calls, evals before prompt-tweaking, and prompt-injection defense. Use whenever writing code that calls an LLM API, designing agent tools or MCP servers, building agents or multi-agent pipelines, writing or tuning prompts, adding RAG/embeddings, or debugging why an LLM feature behaves badly.
prompt-engineer
gustavogutierrez/engineering-skills
Design and engineer System Prompts, prompt templates, and multi-agent orchestration contracts for deterministic, leak-proof AI systems. Use when creating agents, writing skill definitions, designing prompt templates with safe variable injection, structuring I/O contracts, or building multi-agent pipelines.
fine-tuning
pluginagentmarketplace/custom-plugin-ai-engineer
LLM fine-tuning with LoRA, QLoRA, and instruction tuning for domain adaptation.
agentic-workflow
usermeme/skills
Operating discipline for autonomous coding agents, modeled on how Claude Code / Fable 5 works — verify before assuming, clarify with structured questions, challenge flawed plans, plan-then-approve, track tasks visibly, parallelize independent work, verify end-to-end before claiming done, communicate outcome-first, and stay safe around destructive actions. Use at the start of ANY non-trivial engineering task (new feature, refactor, bug hunt, multi-file change), whenever requirements are ambiguous or user-provided "facts" need verification, whenever a proposed approach looks flawed, and whenever you are about to report work as finished. If a task will take more than a couple of steps, this skill applies.
ai-accelerated-building-ng
jona/ycombinator-skills
Apply Andrew Ng's startup building principles and AI-accelerated development strategies from AI Fund's experience launching ~1 startup per month. Use when users ask about startup execution speed, AI coding tools for faster prototyping, agentic AI workflows, evaluating AI startup opportunities, or building AI applications. Triggers include questions about how to build startups faster, AI technology stack layers, where AI opportunities exist, implementing agentic workflows, or applying lessons from successful AI venture studios.
building-finetuning-datasets
osmontero/opencode-skills
Use when preparing to fine-tune, post-train, distill, or LoRA/QLoRA-tune an LLM, when generating synthetic or instruction training data, building SFT/DPO/preference sets, reasoning traces or tool-call trajectories, when choosing rank/alpha/learning rate, when deciding fine-tuning vs RAG, or when a tuned model regressed — hallucinates, lost general ability, repeats, or ignores its format.
ai-agent
vertikon/eduuw-skills
Configure the eduuw AI bot and keyword auto-replies over inbound WhatsApp.
managing-path-cleaning-rules
posthog/ai-plugin
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 path-cleaning-rules-update MCP tool.