mcp skills

14,319 tagged mcp, measured the same way as everything else here.

Browse within: LLM 344agents 269mcp-server 223antigravity 191agent 190agentic-skills 190claude-code-plugin 189ai-workflows 186sdk 168azure 167foundry 167claude-plugin 141agent-marketplace 139skills 116

callput-option-trader

1729

ayggdrasil/options_trading_base

Skill Claude CodeCodex

Compact Callput trading skill for Base L2. Enforce spread-only execution with strict validation, liquidity checks, and close/settle lifecycle rules.

not rated 1 6mo ago A 36 tokens original MIT archived

perch

1731

adityaarsharma/perch

Skill Claude Code

Server intelligence layer for RunCloud-managed Linux servers. Use when the user mentions Perch, /perch, RunCloud, nginx-rc, server intelligence, server diagnosis, WordPress site auditing, plugin vulnerability scanning, or any time they want to investigate, monitor, or heal a server they own.

not rated 1 4mo ago A 65 tokens

zxkol

1733

kvalen-code/zxkol-skill

Skill Claude CodeCodex

A conversational search tool for social-media creators, popular posts, trending topics, and audience sentiment across platforms such as TikTok, YouTube, Instagram, and Xiaohongshu.

not rated 1 2mo ago A 90 tokens original MIT

sanban

1734

aancw/sanban

Skill Claude CodeCodex

Simple kanban that just works. JSON-backed boards with REST API + MCP server.

not rated 1 2mo ago A 20 tokens original MIT

codex-conductor

1735

yksanjo/codex-conductor

Skill Codex

Situational awareness across local Codex sessions. Use when the user says "sort out my Codex sessions", "what is Codex doing", "summarize my Codex windows", "codex conductor", or "what's running".

not rated 1 2mo ago A 54 tokens original MIT

unified-search-mcp

1736

cn-hash/unified-search-MCP-Server

Skill Claude CodeCodex

A search and website-analysis server for Douyin, Xiaohongshu, Zhihu, and CSDN, four Chinese video, social, question-and-answer, and technical-article platforms. It can also inspect a website’s APIs and network requests and analyze how its request signatures work.

not rated 1 2mo ago A 25 tokens original MIT

rag-paper-query

1737

i1ight/ragPaper

Skill Claude CodeCodex

Use when the user asks to search, inspect, summarize, compare, cite, or reason over locally indexed paper PDFs in rag-paper via MCP tools or direct local Chroma/rag-paper access.

not rated 1 1mo ago A 43 tokens original MIT

social-dive

1739

krishddd/social_dive

Skill Claude CodeCodex

Read and search 15+ research/dev/web sources (arXiv, GitHub, PubMed, Semantic Scholar, YouTube, Wikipedia, Hacker News, Stack Overflow, DEV.to, RSS feeds, Crossref, OpenAlex, Europe PMC, DOI resolver, and any web page). USE THIS whenever the user wants to research a topic, read a paper / repo / article / video…

not rated 1 2mo ago A 124 tokens original MIT

geo-audit-mcp

1740

carter-wzq/geo-audit-mcp

Skill Claude CodeCodex

Runs a comprehensive GEO (Generative Engine Optimization) audit on a single URL. Checks 31 signals across 6 pillars and returns a score, pillar breakdowns, per-check pass/fail/warn results, and an actionable fix prompt.

not rated 1 2mo ago A 0 tokens original MIT

meal-planning

1741

LooseWireDev/mealforge

Skill Claude CodeCodex

Plan meals collaboratively in chat, then publish the finalized plan to mealforge via its MCP tools. Use when the user wants to plan meals (a week of dinners, a few breakfasts, a single meal), revise a pushed plan, switch or complete the active plan, or push a plan to mealforge.

not rated 1 2mo ago A 65 tokens original MIT

publish-artifact

1742

saidmukhamad/serve-mcp

Skill Claude Code

Publish HTML, Markdown, CSV, JSON, or a folder to the local serve-mcp artifact shelf and give the user a browser URL. Use whenever you produce a report, chart, diagram, dashboard, mockup, static site, or any file the user would want to view in a browser — without being asked.

not rated 1 13d ago A 68 tokens original MIT

agent-vitals

1743

anirudhprashant/agent-vitals

Skill Claude CodeCodex

Introspect the user's local AI agent infrastructure in real time. Use proactively: before starting tasks, before scheduling cron/timer work, when stuck, after long tasks, or when claiming something works. Exposes tools: vitalssummary, shadowlist, shadowstale, burnoutsummary, burnoutstucksessions.

not rated 1 2mo ago A 68 tokens

skills

1744

chenshuai9101/agent-fde

Skill Claude CodeCodex

Instructions for using an agent that guides an enterprise AI project through discovery, assessment, architecture, prototype, and handoff stages. The process produces files and other structured deliverables, with analysis supplied by the host agent.

not rated 1 1mo ago A 0 tokens original MIT

renderdoc-capture

1745

dcc-mcp/dcc-mcp-renderdoc

Skill Claude Code

Domain skill — Discover RenderDoc and launch an explicit game or graphics executable under capture. Use for repeatable local or CI frame capture. Not for reading an existing capture — use renderdoc-analysis.

not rated 1 20d ago A 44 tokens original MIT

grok-membership-media

1746

Zhao73/grok-membership-media-mcp

Skill Claude CodeCodex

Use the local paid ChatGPT and Grok memberships to create website first frames and real MP4 video without developer APIs or API keys.

not rated 1 1mo ago A 33 tokens original MIT

vibe-blueprint

1747

SinboBoop/vibe-blueprint

Skill Codex

Turn a project idea into a clear technical starting blueprint. Make sure to use this Skill whenever the user is starting a new, non-simple Web, SaaS, internal-tool, or AI Web project, even if they only describe the idea and do not explicitly ask for a tech stack; also use it when choosing or changing architecture or a…

not rated 1 4d ago A 123 tokens original MIT

shared-margin

1748

idleprocesscc/shared-margin

Skill Codex

Read books or watch films together with the user through a linear, spoiler-safe GPT-5.6 workflow that preserves progress, anchored human and model notes, open questions, and a durable trail of shared moments. Use when the user asks to co-read, continue a book, co-watch or analyze a film scene-by-scene, respond to…

not rated 1 1mo ago A 93 tokens original MIT

cursor-skill

1749

WangYuecheng1106/Magic-Mouse

Skill Claude CodeCodex

A computer-use skill for controlling a desktop through Magic Mouse, a desktop pet that provides screen interaction. It can inspect screenshots, plan actions, click, type, switch applications, and verify results.

not rated 1 1mo ago C 0 tokens original MIT

trailmem

1750

amitnexTech/trailmem

Skill Claude CodeCodex

Use when saving or recalling persistent memory via trailmem MCP tools (trailmemstore, trailmemquery, trailmemwelcome, ...), when a trailmem tool call fails or is rejected, or when unsure how to fill its parameters (project scoping, agenttype, eventtype, dedup, linking, archiving).

not rated 1 1mo ago A 69 tokens original MIT

Prove Before Act

1751

jasonxkensei/prove-before-act

Skill Claude CodeCodex

Prove Before Act proof and accountability integration for AI agents. REST API, MCP, and x402. Check live endpoints for availability, pricing, and proof status.

not rated 1 24d ago A 38 tokens

noosxe/worker-mcp

Skill Claude CodeCodex

Guides the coordinator agent on how to orchestrate, delegate to, and supervise local worker agents using the worker-mcp server.

not rated 1 3d ago A 33 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: