skill-security-review
An agent skill that performs structured security reviews of automation skills, deployment plans, and live setups -- finding real risks like credential exposure, injection vectors, blast radius gaps, and failure cascading.
7,504 个自动化任务的工具和技能
An agent skill that performs structured security reviews of automation skills, deployment plans, and live setups -- finding real risks like credential exposure, injection vectors, blast radius gaps, and failure cascading.
ExaAiAgent — Advanced AI-powered penetration testing framework with Docker sandbox, multi-agent workflows, and 50+ integrated cybersecurity tools.
Strava from your terminal. Pipe it, script it, automate it.
AI-powered feature development orchestrator — PRD → Tech Spec → Tasks → Implementation → Tests → PR — with checkpoint/resume, 5 execution modes, and auto-detection for GitHub/GitLab/Azure DevOps. Claude Code skill.
Skill n8n pour Paperclip AI — by Dr. FIRAS
🧠 Five-layer memory system for OpenClaw agents — persistent, cross-session, cross-platform context that never forgets
Full Tesla control from the terminal - dashboard, scheduled preconditioning, energy monitoring, charge station finder, multi-vehicle management via Fleet API. OpenClaw skill.
Fast CLI for browser automation — Playwright behind a persistent daemon
Троица — multi-model decision workflow для Hermes: Conductor (GPT) + Worker (DeepSeek) + Critic (Gemini)
Codex skill for VRChat avatar modding workflows in Unity
Project governance for AI-assisted development. Structured planning, sprint workflows, decision tracking, and scope enforcement — built as plain markdown files and slash commands on top of Claude Code.
Ant-inspired modular, extensible coding workspace designed for AI development workflows. Codomyrmex integrates tools for building, documenting, analyzing, executing, and visualizing code across multiple languages.
A Claude skill that builds your daily schedule around your biology — circadian rhythm, dopamine regulation, and attention science.
Skills and C# command recipes for AI agents controlling the Unity Editor via MCP
A Claude Code meta-skill for creating automated research repositories with paper discovery, classification, and documentation workflows. Works for any research domain.
Agent skill for spec-driven development with AI coding agents — structured specification-first workflow (specify → plan → tasks → implement → validate) for Claude Code, Cursor, GitHub Copilot, and similar tools.
Behave as a Senior QA Engineer covering the complete software testing lifecycle from requirements analysis through test automation and continuous improvement. Always orchestrate available skills to determine the best approach before executing tasks.
A coding-agent skill for safer AI code editing: inspect first, patch minimally, verify honestly.
Claude Code skill: general-purpose GitHub PR review workflow via gh CLI — multiple review styles, optional PR comment posting
Enforce high-rigor completion workflows for coding, UI, copy, and specification-driven tasks. Use when the user says to be thorough, invokes $be-thorough, asks for exhaustive/rigorous implementation, requests no shortcuts, wants meticulous UI or text review, or gives a comprehensive prompt, PRD, audit report, design document, checklist, issue list, or similar source of truth that must be fully satisfied before finalizing.
AI-powered UI test automation — Record browser flows, convert to Playwright + Cucumber BDD tests using Claude Sonnet and Chrome DevTools MCP
AI Agent Skill for NSAuditor AI — gives any AI coding agent built-in knowledge of NSAuditor's MCP tools, schemas, plugins, and security audit workflows. Works with Claude Code, Cursor, Windsurf, and any MCP-aware agent.
A lightweight Git workflow built around main, feat/*, and automatic worktrees.
Agent skill for bounded goal loops, multi-agent review, adaptive sessions, and safe implementation variants.
Calendly scheduling integration. List events, check availability, manage meetings via Calendly API.
Business dashboard that tracks your MRR, clients, pipeline, outreach, and content in one place. Connects to Stripe and Instagram for live data. Set up interactively through a Claude Code skill — no coding required
Agent-agnostic git worktree skill for monorepos. Creates isolated worktrees with recursive env/secret file copying and node_modules symlinking. Supports create/delete/merge workflow.
Open source maintainer automation skills for Claude Code. Issue triage, PR review, releases, health checks — all from slash commands.
>
Standardized multi-agent orchestration framework for coding projects using parallel processing across flagship AI models, preferably local-first.
External Babysitter orchestration for Codex with Windows/Linux install docs, full long-session scenario testing (3+ interview breakpoints, choice validation, strict 100 score), and compat-core fallback.
Build and edit Dify workflow DSL files. Use when creating new Dify workflows from scratch, modifying existing workflows (adding/removing nodes, changing connections), validating workflow structure, or designing LLM-based automation flows with code execution, conditional logic, loops, and error handling.
Office Automation Skill for OpenClaw - Automate Word and Excel processing
Mandatory pre-push security & quality pipeline for Claude Code. Blocks hardcoded credentials, runs parallel review agents, enforces OWASP Top 10.
AI DevOps toolkit for AI-assisted development. Release pipeline, license compliance, repo management, identity protection, best practices.
Claude Code skill pack: hooks, guardrails, and git workflow automation for any dev project. Includes /bootstrap and /audit-project commands.
文章分析工作流自动化 Skill - 安全优化版
Claude Code skill for generating Rewst workflow specs for RoboRewsty
AI skill generator powered by NotebookLM MCP + Claude pipeline. Creates reusable SKILL.md files following skill-creator standards. Supports auto mode (requires notebooklm-mcp) and manual mode.
Codex skill for deleting synced Microsoft Edge passwords through the official UI
Claude Code skills for .faf — reusable workflows for project DNA management
HiveBoard is a local-first Kanban board for AI agent orchestration. You create tasks on the board UI, and HiveBoard dispatches autonomous coding agents (Claude CLI) to complete them
Build a LaTeX Beamer presentation strictly from the bundled minimal Beamer template (assets/template/). Use this skill whenever the user mentions making slides, a presentation, a deck, a talk, or a Beamer/LaTeX presentation — in English ("slides", "deck", "talk", "presentation", "make me a deck about X"), 繁體中文 (「投影片」「簡報」「報告投影片」「做一份簡報」「做投影片」「做 PPT」「做 LaTeX 簡報」「Beamer 簡報」), or 简体中文 (「幻灯片」「演示文稿」「做一份演示」「做演示文稿」「做幻灯片」「做 PPT」「做 LaTeX 演示」「Beamer 演示」) — even if they just dump a topic, a paper, or some notes. Treat "PPT" / "ppt" as a request for slides regardless of whether the user actually wants a .pptx (default to Beamer; only switch if they explicitly say PowerPoint). Do NOT start writing .tex without running the clarification workflow described in this skill first; vague inputs (a bare topic word, "做一份关于 diffusion models 的简报", "做我硕士论文的 defense slides", "make a deck about X") MUST go through clarification before any frame is drafted. After every build, a mandatory revision loop engages — the number of follow-up rounds is driven by the user's feedback, not capped. Use this skill instead of inventing a Beamer setup from scratch — the template is the source of truth and must not be replaced or restyled.
Disciplined agentic development for Claude Code. Auto-routing orchestrator with XML specs, atomic commits, multi-stage review pipeline, and zero slop.
headless high performance query engine on markdown files that can integrate into various workflows
🤖 AI-powered Terraform module management across AWS, GCP, Azure, DigitalOcean - Save 97% of maintenance time
Obsidian vault generator, automation daemon, and graph strengthener
Autonomous triangular arbitrage scanner and AI trading assistant for Binance. Combines all 7 official Binance AI Agent Skills into a unified HFT-grade workflow with real-time triangle detection, spoof hunting, risk management, and natural language control. Built for the Binance OpenClaw Challenge.
Concept-first writing pipeline that drafts, evaluates, and revises in fixed order (concept → structure → line). Bring your own voices.
A Claude Code skill that orchestrates a 5-tier agent pipeline to produce comprehensive implementation plans for complex features