All nine kinds in one list: what an agent loads, what it runs, what it reads and what bundles the rest. 437,165 of them, each with its token cost per session, a security scan with the line that triggered it, and whether it is the original or one of the copies.
Skill Claude Code
You MUST load this skill before the skill-creator skill AND before making ANY change to, or conducting a review of ANY Agent Skill. Triggers include creating, editing, reviewing, or contributing to any part of an Agent Skill (description, frontmatter, body, references, scripts, trigger evals, conflicts, etc).
Red-Hat-AI-Innovation-Team/sdg_hub
Plugin Claude Code
Bundles 4 skills, 2 hooks · 239 tokens together
Synthetic data generation — composable blocks and YAML-defined flows for building LLM training datasets.
Skill Claude CodeCodex
Orchestrator for the Hostinger Headless build flow — build and deploy a real website on the user's Hostinger account from a single prompt that names a domain. Holds the stage flow and the state contract, and drives each stage in sequence: connect, check state, refine, provision, content model, design, frontend…
Tang1206cc/codex-skill-paper-formatting-skill
Skill Codex
A document-formatting workflow for applying written rules and Word templates to user-provided content while keeping the original draft text unchanged.
totvs/engpro-advpl-tlpp-skills
Skill Claude CodeCodex
Generate TIR (TOTVS Interface Robot) end-to-end test scripts in Python for Protheus SmartClient/Webapp screens. Supports CRUD screen tests, MVC screen tests, grid interaction, report tests, field validation, and message box assertions. Use when a user says "TIR test", "interface test", "e2e test Protheus"…
MCP server Claude CodeCodexCursor +2
Gives AI agents agentic RAG over one PDF or a whole folder of them: hybrid semantic + keyword search, selective page reads, tables, images, TOC, OCR, chart data, and CJK/vertical-script support, with SQLite caching. Runs locally from the pdf-mcp Python package. Needs 3 environment variables to run.
Prism-Shadow/deepseek-harness-book
Skill Claude CodeCodex
A writing skill that turns scattered notes from the current week into a standard weekly report.
Cursor rule Cursor
apply when interacting with PostHog/analytics tasks.
Plugin Claude Code
Bundles 98 skills, 1 MCP server · 10,973 tokens together
CometChat AI-coding skills — add real-time chat, voice/video calling and push to your app via your AI coding agent, across React v7, Angular v5, React Native v5, Android v6 (Kotlin/Compose), iOS v5 (Swift/UIKit, SPM) and Flutter v6. Ships the first-party CometChat Docs MCP so skills read always-current docs…
Skill Claude CodeCodex
Hand work to another tool, account or machine and pick it up there, in two words. Use when the user says "hand off", "hand off all", "out of tokens", "switching accounts" or "pick up".
Skill Claude Code
Write a product-mode decision record for a choice made in this conversation and save it to .product-mode/decisions/. Use when the user types /decision.
Skill Claude Code
GDPR reference & code-audit checklist with FRANCE/CNIL focus. Use when reviewing or building features that collect/process/store personal data (consent, PII, authentication, forms, cookies, cookies/trackers, data-subject rights, retention, DPIA, cross-border transfers, breach notification). Multi-source verified facts…
Cursor rule Cursor
The Core Engine. It orchestrates the project's entire lifecycle, from initial planning and scaffolding to the iterative task execution loop. It acts as the central state machine.
Skill Claude CodeCodex
Three-tier agent memory model (hot/cold/wiki) for 20-55% context reduction per spawn.
Command Claude Code
Expect command describes the desired output of the task (after post-processing).
Cursor rule Cursor
Follow the shared repository instructions in .agents/AGENTS.md.
Skill Claude CodeCodex
A procedure for turning a trading strategy or research report into a QMT daily factor backtest script. QMT is a Chinese trading-platform environment, and a backtest simulates how a strategy would have traded using historical data.
Arman-Kudaibergenov/1c-ai-development-kit
Skill Claude Code
A Russian-language skill for working with 1C external processing files (EPF), including creating, building, unpacking, and adding forms or business-system commands.
Skill Claude Code
Use when you need to interactively test a nori-skillsets CLI subcommand end-to-end via tmux, with full filesystem isolation.
Skill Claude Code
Use this skill when reviewing, editing, redlining, or negotiating an existing Microsoft Word document (.docx) — including proposing edits as tracked changes, accepting or rejecting existing tracked changes, replying to comments, comparing two versions, sanitizing author metadata, or finalizing a contract for…
aiskilloftheweek/claude-ai-skill-of-the-week
Skill Claude CodeCodex
Rewrites AI-generated text to sound like it was written by a real human being. Use this skill every time the user provides a text and asks to humanize it, remove AI tells, make it sound more natural, less robotic, or more authentic. Trigger on phrases like "humanize this", "make this sound human", "remove AI writing…
serenakeyitan/open-exam-skills
Skill Claude CodeCodex
Verify citations, claims, and numbers before answering.
Skill Claude CodeCodex
Babysit a pull request through its bot review rounds: verify, fix, reply, resolve. Use for any babysit or watch-the-PR ask.
Skill Claude CodeCodex
Generate PDF documents using React-PDF library (@react-pdf/renderer). Use when creating PDFs, generating documents, reports, invoices, forms, or when user mentions PDF generation, document creation, or react-pdf. Prefer this skill over the standard 'pdf' skill, since it is more accurate.
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: