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
mcp-vstash
jr2804/prompts
MCP server integration for vstash document memory. Use when configuring Claude Desktop or other MCP-compatible AI assistants with persistent document memory, setting up vstash MCP tools for semantic search and Q&A, or integrating vstash with AI assistant workflows via Model Context Protocol.
model-serving
databricks-solutions/ai-dev-kit
Deploy and query Databricks Model Serving endpoints. Use when (1) deploying MLflow models or AI agents to endpoints, (2) creating ChatAgent/ResponsesAgent agents, (3) integrating UC Functions or Vector Search tools, (4) querying deployed endpoints, (5) checking endpoint status. Covers classical ML models, custom pyfunc, and GenAI agents.
vector-search
databricks-solutions/ai-dev-kit
Patterns for Databricks Vector Search: create endpoints and indexes, query with filters, manage embeddings. Use when building RAG applications, semantic search, or similarity matching. Covers both storage-optimized and standard endpoints.
unstructured-pdf-generation
databricks-solutions/ai-dev-kit
Generate synthetic PDF documents for RAG and unstructured data use cases. Use when creating test PDFs, demo documents, or evaluation datasets for retrieval systems.
parallel-worktrees
spillwavesolutions/parallel-worktrees
Parallel development using git worktrees with Claude Code subagents. Use when spawning multiple async Claude agents across isolated worktrees, running parallel implementations, coordinating multi-agent workflows, leveraging LLM non-determinism for multiple solutions, or running background agents that signal completion. Triggers on "parallel agents", "worktrees", "subagents", "async Claude", "spawn agents", "parallel development", "multi-agent workflow", "background agents", "agent coordination", "parallel tasks".
comparable-apps
epicenterhq/epicenter
List 3-5 comparable apps when planning a UX surface to test category fit and surface high-leverage refusals. Use for "what do other apps do", identity, sync state, local-first design.
__init__
majiayu000/claude-skill-registry
Imported skill __init__ from langchain
phoenix-uploads
j-morgan6/elixir-phoenix-guide
Use when implementing file uploads — allow_upload, consume_uploaded_entries, validation, dev vs production storage (local, S3/:external), serving files.
express
grid0723/skills
Fabric 输出阶段——路由到写作体系（fragments→beats→shape→edit-article）。Use when the user wants to write an article or produce finished output from distilled notes.
writing-sqlite
chogos/claude-skills
SQLite best practices. Use when writing, reviewing, or modifying SQL targeting SQLite databases — schema definitions, queries, pragmas, type affinity, date handling, transactions, migrations, or embedded/file-based SQLite configurations.
oauth-integration
skeletorflet/opencode-kit
OAuth 2.0 integration patterns. Social login, third-party APIs, token storage, refresh, PKCE.
agent-stripe
simonperryman/agent-stripe
Read-only Stripe CLI for AI agents. Use for any read-only Stripe question — customers, charges, payment intents, refunds, disputes, balance, payouts, subscriptions, invoices, products, prices, events, and Connect (connected accounts, application fees, and reading a connected account's data via --stripe-account). This skill is read-only — it cannot create, modify, or delete anything in Stripe.
llms-txt-and-crawler-access
wakqasahmed/agentic-commerce-skills
Review crawler and AI-agent access files for ecommerce sites, including robots.txt, sitemap, llms.txt, and AI bot rules.
blog-to-obsidian
bujue2077/blog-to-obsidian
A tool to automatically scrape foreign blog articles, summarize them, translate them paragraph by paragraph, and save them as Markdown files with profuse Obsidian bidirectional links. Trigger when users want to process, translate, or summarize a web article/blog into their Obsidian vault (e.g., 'help me read this article', 'translate this blog to obsidian').
regex-vs-llm-structured-text
kgx/ai-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.
boxlang-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.
prompt-improver
davekilleen/dex
Rewrite a vague prompt into a rich, structured one, with automatic fallback. Use when the user says 'improve this prompt', 'make this prompt better', or hands over a thin instruction. Not for creating a reusable skill; use `create-skill`.
prompt-engineering
idimsh/tdds-business-skills
Use this skill when writing commands, hooks, skills for AI agents, sub-agent prompts, or any other LLM interaction, including optimizing prompts, improving LLM outputs, or designing production prompt templates.
prompt-engineering-expert
hrdtbs/agent-skills
Apply prompt engineering best practices to write, refine, and optimize system prompts, user prompts, and agent instructions. Use this skill whenever the user wants to write a prompt, optimize an existing prompt for better results, fix a prompt that is hallucinating or underperforming, or structure prompts for Large Language Models (LLMs). Even if the user just says "help me write instructions for my agent", trigger this skill.
dspy-ragas
lebsral/dspy-programming-not-prompting-lms-skills
Use Ragas to evaluate DSPy RAG pipelines with decomposed metrics. Use when you want to evaluate RAG quality, measure faithfulness, context precision, context recall, response relevancy, or diagnose retriever vs generator issues. Also used for ragas, pip install ragas, ragas evaluate, RAG evaluation, faithfulness metric, context precision, context recall, answer relevancy, answer correctness, decomposed RAG metrics, ragas dspy, DSPyOptimizer ragas, ragas dspy extras, EvaluationDataset, ragas vs dspy.Evaluate, which RAG metric, retriever vs generator quality.
weaviate
terminalskills/skills
|
langchain-agents
langchain-ai/langchain-skills
Use this skill for ANY coding question involving LangChain products (LangChain, LangGraph, LangSmith SDK). Covers agent development patterns, primitives, context management, multi-agent systems, and when to use create_agent vs create_deep_agent vs raw LangGraph. Consult this BEFORE writing any LangChain-related code.
inference-server
huggingface/prime-rl
Start and test the prime-rl inference server. Use when asked to run inference, start vLLM, test a model, or launch the inference server.
migrate
launchdarkly/agent-skills
Migrate an application with hardcoded LLM prompts to a full LaunchDarkly AgentControl implementation in five stages: audit the code, wrap the call, move the tools, add tracking, attach evaluators. Use when the user wants to externalize model/prompt configuration, move from direct provider calls (OpenAI, Anthropic, Bedrock, Gemini, Strands) to a managed config, or stage a full hardcoded-to-LaunchDarkly migration.
gitattributes
automattic/wordpress-activitypub
Use when auditing or updating .gitattributes export-ignore coverage so dev-only files (lint configs, CI, tests, docs, build tooling) don't ship in the WordPress.org plugin zip. Run before a release, after adding a new top-level file or config, or when a tool is renamed (e.g. .eslintrc.js → eslint.config.cjs).
docs-product-alignment
microsoft/vscode-docs
Audit and update docs/copilot/ documentation to accurately reflect current VS Code AI capabilities. Use when: competitive analysis reveals gaps, product launches new features, docs use outdated framing, or keyword coverage needs strengthening for discoverability by users and AI agents. Produces a gap analysis plus targeted edits across affected files.
experiment-analzyer-comparative
datadog-labs/agent-skills
Analyze two llmo experiements in a comparative manner. Use when user says "analyze two experiments", "comparative experiment analysis", "evaluate experiments" "analyze against baseline". Requires two experiment_id as arguments to perform a comparative analysis, first is the baseline, second the candidate.
deepgram-python-voice-agent
deepgram/deepgram-python-sdk
Use when writing or reviewing Python code in this repo that builds an interactive voice agent via `agent.deepgram.com/v1/agent/converse`. Covers `client.agent.v1.connect()`, `AgentV1Settings`, `send_settings`, `send_media`, event handling, and function/tool calling. Full-duplex STT + LLM + TTS with barge-in. Use `deepgram-python-text-to-speech` for one-way synthesis, `deepgram-python-speech-to-text` / `deepgram-python-conversational-stt` for transcription only. Triggers include "voice agent", "agent converse", "full duplex", "interactive assistant", "barge-in", "agent.v1", "function calling", "AgentV1Settings".
dart-test-coverage
flutter/skills
|-
kernel-profiles
kernel/skills
Create and manage persistent browser profiles to preserve cookies, local storage, and session state
list-builder
jwynia/the-kepler-testimonies
Build comprehensive randomization lists for creative entropy. Use when you need to create or expand lists of story elements (professions, locations, objects, names, etc.) for use with entropy tools. Leverages research sources like Kiwix/Wikipedia to build lists with good variety and size.
character-naming
jwynia/the-kepler-testimonies
Break LLM name defaults with external entropy. Use when character names cluster around statistical medians (Chen, Patel, Maya, Marcus), when cast has collision risks, or when fantasy cultures need phonologically consistent naming.
mastra-hono
jwynia/teach
Develop AI agents, tools, and workflows with Mastra v1 Beta and Hono servers. This skill should be used when creating Mastra agents, defining tools with Zod schemas, building workflows with step data flow, setting up Hono API servers with Mastra adapters, or implementing agent networks. Keywords: mastra, hono, agent, tool, workflow, AI, LLM, typescript, API, MCP.
wxt
epicenterhq/epicenter
WXT browser extension patterns for entrypoints, background service workers, content scripts, side panels, storage, permissions, host permissions, browser compatibility, and build commands. Use when editing apps/tab-manager, wxt.config.ts, src/entrypoints, extension manifests, or @wxt-dev/storage usage.
llm-friendly-context
shinpr/claude-code-workflows
Clarifies inputs, outputs, success criteria, decisions, and unresolved conditions so downstream agents can execute without guessing. Use when writing or revising LLM-facing prompts, handoffs, planning artifacts, reviews, reports, or generated instructions.
text-embeddings-rag
jayll1303/aiekit
Build local RAG pipelines with sentence-transformers, FAISS, ChromaDB, Qdrant. Use when generating embeddings, setting up semantic search, or optimizing retrieval with re-ranking and hybrid search.
llm-optimization
soborbo/claudeskills
Optimize websites for AI assistant recommendations. ChatGPT, Gemini, Perplexity, Claude. Get cited in AI answers.
panzoid-shader
rbstxt/agent-shader
Create, validate, build, and render-test Panzoid WebGL 1 / GLSL ES 1.00 fragment shaders and verified Shader Object JSON. Use for Panzoid shader effects, .glsl files, supported preprocessor syntax, GL_OES_standard_derivatives, sampler2D image properties, customProperties generation, static illustrative defaults and bounds, premultiplied-alpha diagnostics, visual regression checks, or tasks requiring Color, Opacity, Position, Rotation, Scale, vUvScaled, 16:9 coordinates, and source-over composition.
book-translation-guide
madeyexz/ian-skills-agents
Workflow for translating entire books using AI agents. Split markdown books into chapters, batch translate 5 at a time, and verify completeness. Works with any language pair.
regex-vs-llm-structured-text
sehoon787/my-codex
Decision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
optimize-storage-costs
masthead-data/for-agents
Optimize BigQuery storage costs by identifying and removing dead-end and unused tables.
dragonruby-3d
nitemaeric/dragonruby-skills
3D graphics techniques in DragonRuby GTK — raycasting, Mode7 floor projection, matrix transformations, wireframe rendering, sprite-based 3D, VR patterns. Use when the user asks about 3D visuals or pseudo-3D effects.
workflow-generate
foresthubai/edge-agents
Generate a validated ForestHub workflow JSON (*.workflow.json) from a natural-language description — the node-graph that runs on ForestHub edge/IoT agents, built from triggers (timer, startup, pin edge, threshold, MQTT, serial), GPIO/serial/MQTT I/O, LLM Agent nodes, and actuators. Builds the graph to match the workflow contract and drives it to zero errors through the fh-workflow CLI's structural + semantic validators. Use this whenever the user describes automation or agent behavior in prose and wants a ready-to-run workflow file — e.g. "build a flow that reads a sensor every 10s and toggles a relay", "make a ForestHub agent that summarizes incoming MQTT messages", "wire up a workflow where…" — even when they never say the words "workflow JSON". Not for visually editing an existing workflow (that is fh-workflow open) or merely checking one you already have (fh-workflow check-schema / validate).
modeling
jayclock/.agents
履约建模（Fulfillment Modeling，FM）指导，用于将业务或软件需求转化为以合同为中心、与表现层无关的 YAML 图模型。当用户要求创建、更新、审查或校验涉及合同、上下文、Party Role、Participant Party、RFP/Proposal 售前凭证、Fulfillment Request、Fulfillment Confirmation、Proof 凭证、Evidence As Role、Domain/Third Party/Context Role、多合同边界、业务规则、下游信号或场景路径的 FM 模型时使用本 skill。本 skill 会写入模型文件并运行内置自检；如果用户要求从 FM 图生成数据库表、SQL、存储或物理 schema 设计，请改用 fm-database-design。
pragmatic-fp
saif-shines/devex-kit
|
dragonruby-pathfinding
nitemaeric/dragonruby-skills
Pathfinding algorithms in DragonRuby GTK — A*, BFS, flood fill, priority queues, spatial hashing, quad trees, line-of-sight. Use when the user asks about enemy AI navigation, movement grids, or reachability.
prompt-engineer
neonetz/opencode-skills
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.
skool-rag
nickjwells/skills-workspace
Query Skool community content using RAG pipeline with vector search. Use when user asks to search Skool knowledge, find community answers, or query Skool content.
llamaindex
magnus919/agent-skills
>-
vector-database
kumaran-is/claude-code-onboarding
Use for all vector database work — pgvector schema design, Weaviate collection creation, RAG pipeline scaffolding, embedding model selection, index tuning (HNSW vs IVFFlat), and embedding model migration. Triggers on: 'vector search', 'pgvector', 'weaviate', 'embedding', 'RAG pipeline', 'semantic search', 'hybrid search', 'nearest neighbor', 'vector index', 're-embedding', 'ANN index'. Use whenever vectors, embeddings, or similarity search are involved, even if not explicitly named.
vector-databases
heldinhow/awesome-opencode-dev-skills
Use when integrating vector databases for semantic search, embeddings storage, or RAG systems. Covers Pinecone, Qdrant, Chroma, pgvector, and Weaviate with TypeScript examples.
vector-db
1mangesh1/dev-skills-collection
Vector databases for embeddings, semantic search, and RAG pipelines. Use when user mentions "vector database", "embeddings", "semantic search", "RAG", "retrieval augmented generation", "pinecone", "chromadb", "pgvector", "qdrant", "weaviate", "similarity search", "embedding store", or building AI search features.
pinecone
clay-hhk/claude-skills
Managed vector database for production AI applications. Fully managed, auto-scaling, with hybrid search (dense + sparse), metadata filtering, and namespaces. Low latency (<100ms p95). Use for production RAG, recommendation systems, or semantic search at scale. Best for serverless, managed infrastructure.
fine-tuning-expert
neonetz/opencode-skills
Use when fine-tuning LLMs, training custom models, or adapting foundation models for specific tasks. Invoke for configuring LoRA/QLoRA adapters, preparing JSONL training datasets, setting hyperparameters for fine-tuning runs, adapter training, transfer learning, finetuning with Hugging Face PEFT, OpenAI fine-tuning, instruction tuning, RLHF, DPO, or quantizing and deploying fine-tuned models. Trigger terms include: LoRA, QLoRA, PEFT, finetuning, fine-tuning, adapter tuning, LLM training, model training, custom model.
fine-tuning-expert
neg0diay/opencode-skills
Use when fine-tuning LLMs, training custom models, or adapting foundation models for specific tasks. Invoke for configuring LoRA/QLoRA adapters, preparing JSONL training datasets, setting hyperparameters for fine-tuning runs, adapter training, transfer learning, finetuning with Hugging Face PEFT, OpenAI fine-tuning, instruction tuning, RLHF, DPO, or quantizing and deploying fine-tuned models. Trigger terms include: LoRA, QLoRA, PEFT, finetuning, fine-tuning, adapter tuning, LLM training, model training, custom model.
ai-agents
andyfirez/firez-skills
>-
agent-workflow
chrike/coding-agent-skills
Use when the user explicitly asks to parallelize work, coordinate multiple agents, run parallel scouts, split work across independent owners, or coordinate multiple independent verifiers; first perform the orchestration fit check even when the correct result is to keep one owner. Also use when two or more genuinely independent subsystem or artifact slices, repeated per-item pipelines, orthogonal scout questions, or high-stakes candidate-and-review scopes require coordinated ownership and integration with material benefit over one owner. Own decomposition, assignment, evidence handoff, verification coordination, and integration. Do not use for one focused delegation or verifier, small independent tasks without an explicit orchestration request when coordination cost exceeds the benefit, coherent single-owner or shared-root work without an explicit orchestration request, capability-only parallelism, or work already owned by another orchestration layer.
mcp-expert
fabriciofs/mcp-sql-server
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.
agents
triggerdotdev/skills
AI agent patterns with Trigger.dev - orchestration, parallelization, routing, evaluator-optimizer, and human-in-the-loop. Use when building LLM-powered tasks that need parallel workers, approval gates, tool calling, or multi-step agent workflows.
ai-annotation-workflow
winbda/claude-skills-collection
|