Agent skills

AI & Agents skills

Read straight from the source repositories, not from submitted listings. Every skill shows what it does, what is inside, where it came from — and whether attention around its source is actually growing.

Toolclaude-code 29,140codex 4,755cursor 3,111copilot 976windsurf 55cline 34
CategoryWorkflow & Productivity 4,979AI & Agents 3,037Data & Analytics 2,345Code Review & Quality 1,376Backend & API 1,244Security 1,194Design & Presentation 1,154Documentation 965Content & Marketing 916Testing & QA 777DevOps & Cloud 576Databases 550Frontend 469Business & Finance 328Media & Video 257Other 9,833
4,965 found
2,209–2,256 · page 47 / 104
rag-pipeline-builderDesigns retrieval-augmented generation pipelines for document-based AI assistants. Includes chunking strategies, metadata…majiayu000rag-queryConsulta ao corpus RAG do projeto. Busca conhecimento em decisoes, documentacao, learnings e padroes armazenados. Use quando…majiayu000writesrag-queryConsulta ao corpus RAG do projeto. Busca conhecimento em decisoes, documentacao, learnings e padroes armazenados. Use quando…majiayu000writesrag-query-transformationQuery expansion, HyDE, and multi-query generation for improved retrievala5c-aiwritesrag-retrievalRetrieval-Augmented Generation patterns for grounded LLM responses. Use when building RAG pipelines, constructing context from…majiayu000rag-service高性能 RAG 多路检索服务。集成 Milvus 向量数据库进行语义检索，并结合 Rerank 模型进行精准重排序，支持海量文档的高效存储与历史内容召回。majiayu000rag-skillsRAG-specific best practices for LlamaIndex, ChromaDB, and Celery workers. Covers ingestion, retrieval, embeddings, and…majiayu000rag-system-builderBuild Retrieval-Augmented Generation (RAG) Q&A systems with Claude or OpenAI. Use for creating AI assistants that answer…majiayu000ragdollRagdoll physics skill for joint constraints and hit reactions.a5c-aiwritesraisonnement-juridiqueModélise le raisonnement juridique d'un magistrat français pour l'analyse de dossiers civils. Utiliser ce skill pour analyser un…majiayu000ralphConvert PRDs to prd.json format and run the Ralph autonomous agent system. Use when converting PRDs to Ralph format OR running…majiayu000ralphConvert PRDs to prd.json format for the Ralph autonomous agent system. Use when you have an existing PRD and need to convert it…majiayu000ralphConvert PRDs to prd.json format for the Ralph autonomous agent system. Use when you have an existing PRD and need to convert it…majiayu000ralphML-Ralph autonomous agent. Start ML projects, create PRDs through conversation, run autonomous experiments. Triggers: ralph, ml…majiayu000ralphAutonomous agent loop for completing features. Use when asked to 'use ralph', 'ralph this', or to autonomously implement a…majiayu000ralph-runRun RALPH autonomous development loop with multi-agent pipelinemajiayu000writesralph-setupSet up automated agent-driven development with Ralph. Run AI agents in a loop to implement features from user stories, verify…majiayu000ralph-specum-helpThis skill should be used only when the user explicitly asks to use `$ralph-specum-help`, or explicitly asks Ralph Specum in…majiayu000ralph-specum-researchThis skill should be used only when the user explicitly asks to use `$ralph-specum-research`, or explicitly asks Ralph Specum in…majiayu000ralph-specum-researchThis skill should be used only when the user explicitly asks to use `$ralph-specum-research`, or explicitly asks Ralph Specum in…majiayu000ralph-wiggumAutonomous AI coding loop for unattended development - lets Claude work continuously on tasks while you sleep, with mandatory…majiayu000writesRAN AgentDB Integration SpecialistAgentDB integration specialist for RAN ML systems with vector storage, pattern recognition, and distributed training…majiayu000RAN DSPy Mobility OptimizerDSPy-based mobility optimization with temporal patterns, handover management, and 15% improvement target. Uses program synthesis…majiayu000ray-distributed-trainerDistributed computing skill using Ray for parallel training, hyperparameter search, and resource management.a5c-aiwritesray-distributed-trainerDistributed computing skill using Ray for parallel training, hyperparameter search, and resource management.majiayu000writesraytracingReal-time ray tracing skill for DXR/VK_KHR_ray_tracing and hybrid rendering.a5c-aiwritesrca-human-factorsHuman-factors root cause analysis for incidents involving operators, crews, clinicians, or any human actors. Covers James…majiayu000rca-systems-theoreticSystems-theoretic approaches to root cause analysis for complex socio-technical systems where linear causal chains fail. Covers…majiayu000reExplicit extra-research mode. Use only when the user says `/re` or clearly asks for a research-heavy pass before coding. This…majiayu000read-arxiv-paperUse this skill when when asked to read an arxiv paper given an arxiv URLmajiayu000read-arxiv-paperUse this skill when when asked to read an arxiv paper given an arxiv URLmajiayu000read-arxiv-paperUse this skill when when asked to read an arxiv paper given an arxiv URLmajiayu000readwise-chatChat with Readwise highlights using GPT-5.1 RAG. Use when the user wants to ask questions about their reading library, get…majiayu000reality-check-modeUse for grounding or AI-loop escalation when the user treats ordinary artifacts, numbers, phrases, coincidences, symbols, model…softcanereasoning-modelsMulti-provider reasoning/thinking configuration. Discriminated unions for OpenAI effort, Anthropic budget, Google level/budget…majiayu000reasoning-patterns-v2Use this skill for rigorous theoretical derivation with supercollider mode (G1-G7 simultaneous), diffusion reasoning, and…majiayu000recallRecall prior work from past sessions — how a bug was fixed, what was decided, where a pattern lives. Use when asked 'have we done…oliver-kriskarecap回顾最近 N 天的 Claude Code 使用记录——扫描原始会话数据，按主题分组汇总"我都做了什么"，并从个人操作系统视角输出模式、风险与增删建议。当用户说 /recap、"看看我这几天做了什么"、"回顾一下我最近的会话"、"这两天我用 claude…majiayu000recovering-automation-workflowsUse when recovering automation workflows is required during automation product work, especially when the result must be…casioreview20-glitchrecovering-interrupted-runsUse when recovering interrupted runs is required during kernel work, especially when the result must be traceable, independently…casioreview20-glitchrecsys-pipeline-architectDesigns composable recommendation, ranking, and feed pipelines using the six-stage…majiayu000Recursive DecompositionThis skill should be used when facing long-context tasks, processing large documents, handling multi-file analysis, aggregating…majiayu000reducing-saas-churnUse when reducing saas churn is required during saas product work, especially when the result must be traceable, independently…casioreview20-glitchrefactoring-with-behavior-preservationUse when refactoring with behavior preservation is required during implementation work, especially when the result must be…casioreview20-glitchrefine-promptThis skill should be used when the user asks to "refine a prompt", "optimize a prompt", "improve my prompt", "rewrite prompt for…majiayu000writesregex-vs-llm-structured-textDecision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for…majiayu000Register AllocationExpert skill for register allocation algorithms including graph coloring, linear scan, and spill code generationa5c-aiwritesregression-analysis-modelingPerform comprehensive regression analysis and predictive modeling using linear regression, decision trees, and random forests.…majiayu000writes
← Prev47 / 104Next →
How the catalog works
What is an agent skill?

A folder with a SKILL.md inside — instructions, and often scripts and assets, that an AI agent loads when the task matches. Claude Code, Codex, Cursor and Copilot all read the same format, so one skill usually works across them.

Where does this catalog come from?

We read 660 source repositories straight from their file trees rather than from submitted listings — what you see is what is actually published. 98 repositories were rejected because they advertise skills but contain none: link lists, not folders.

Why is there no install counter?

Because install counts live in the registry that serves `npx skills add`, and that is not ours — publishing a number we cannot verify would be worse than showing none. Instead we show where a skill comes from and whether attention around its source is actually growing, measured from our own weekly snapshots.

Do you deduplicate?

Yes, and it matters more than expected. Aggregator repositories republish the same skill in several places — one source carried 6,317 SKILL.md files for 2,001 actual skills. We collapse by folder name and keep the canonical copy, so the catalog counts things, not copies.

Keep going
Collections, ranked →
Which skill collections are moving this week, from our own snapshots.
Open Source Radar →
The repositories behind these skills, watched daily.
The makers →
Who publishes the skills you install.
Startup rankings →
The companies building in the same ecosystem.