发现 Skills

Unified LLM Gateway - One API for 70+ AI models. Route to GPT, Claude, Gemini, Qwen, Deepseek, Grok and more with a single API key.

This skill enables IP address masking and access to hidden services on the Anyone Network. Route requests through the Anyone Protocol VPN network using a local SOCKS5 proxy.

Extract quantities from BIM/CAD data for cost estimation. Group by type, level, zone. Generate QTO reports.

> **TL;DR**: 本文记录如何通过 Playwright + 浏览器 Session 管理，实现即梦AI视频生成的全自动化。包含图片压缩、Cookie 注入、表单填写、视频提交等完整流程。

Credit card rewards optimizer — helps maximize cashback, points, and miles by recommending the best card for every purchase category. Tracks annual caps, calculates annual fee ROI, manages rotating quarterly categories, and suggests new cards based on spending patterns.

CircleCI CI/CD — manage pipelines, workflows, jobs, and insights via REST API

Modify Clawd, the Claude Code mascot. Use this skill when users want to customize Clawd's appearance in their Claude Code CLI, including changing colors (blue Clawd, green Clawd, holiday themes), adding features (arms, hats, accessories), or creating custom ASCII art variants. Triggers include "change Clawd color", "give Clawd arms", "customize the mascot", "modify Clawd", "make Clawd [color]", or any request to personalize the Claude Code terminal mascot.

Adaptive communication coaching that shapes speaking and writing behavior through reinforcement, scoring, and micro-interventions. Use when the user shares communications for feedback, requests practice scenarios, or during scheduled check-ins. Trains clarity, vocal control, presence, persuasion, emotional regulation, and boundary setting. Based on rhetoric, negotiation, and performance psychology frameworks.

Generates new content drafts based on analysis of reference content. Use this when you want to create content (articles, tweets, posts) modeled on high-performing examples. The tool analyzes reference URLs, extracts repeatable patterns, generates context questions, creates a reusable meta-prompt for LLMs, and produces multiple draft variations.

Create compelling, personalized cover letters that help you stand out in job applications

The first AI religion — a benign memetic experiment in agent network security

name: email-assistant description: Multi-email management assistant supporting Gmail, 163, QQ, Outlook, and Hotmail. Features: (1) Fetch inbox and summarize emails (2) Keyword-based important email...

Endpoints document management API toolkit. Scan documents with AI extraction and organize structured data into categorized endpoints. Use when the user asks to: scan a document, upload a file, list endpoints, inspect endpoint data, check usage stats, create or delete endpoints, get file URLs, or manage document metadata. Requires ENDPOINTS_API_KEY from endpoints.work dashboard.

Create distinctive, production-grade static sites with React, Tailwind CSS, and shadcn/ui — no mockups needed. Generates bold, memorable designs from plain text requirements with an anti-AI-slop aesthetic, mobile-first responsive patterns, and single-file bundling. Use when building landing pages, marketing sites, portfolios, dashboards, or any static web UI. Supports both Vite (pure static) and Next.js (Vercel deploy) workflows.

Write a full-length technical book using multi-agent AI orchestration. Spawns parallel research, writing, and review agents to produce 60K-100K+ word manuscripts. Also supports revising individual chapters of existing books. Based on the real workflow that produced an 88,000-word, 14-chapter book in under 18 hours.

Query and display Juicebox V5 fund access limits (payout limits, surplus allowances). Use when: (1) surplusAllowancesOf or payoutLimitsOf returns empty but values exist, (2) need to detect "unlimited" values which can be various max integers not just uint256, (3) fund access limits show as zero for USDC-based projects when querying with ETH token, (4) REVDeployer stageOf or configurationOf reverts, (5) CONFIGURING project deployment and need unlimited payouts - CRITICAL: empty fundAccessLimitGroups means ZERO payouts not unlimited, must use uint224.max for unlimited. Covers JBFundAccessLimits querying, multi-token support (ETH/USDC), ruleset chain walking, unlimited value detection, and proper configuration for project deployment.

Location-based safety monitoring with automatic alerts and escalation. Use when setting up safety monitoring for a user, tracking their location, checking for nearby dangers (weather, earthquakes, air quality, local emergencies), or configuring emergency contact escalation. Also includes self-preservation monitoring for the agent's host machine. Triggers on requests about safety alerts, location tracking, emergency monitoring, "keep me safe," or "monitor yourself."

Markdown 转 PDF 技能。将 Markdown 文件转换为精美的 PDF 文档，完美支持中文、代码高亮、自定义样式。

Create and manage Microsoft Advertising campaigns (Bing Ads / DuckDuckGo Ads) via MCP server - campaigns, ad groups, keywords, ads, and reporting

Speech-To-Text with MLX (Apple Silicon) and opensource models (default GLM-ASR-Nano-2512) locally.

Security hardening for AI agents - Moltbot, OpenClaw, Cursor, Claude. Lock down gateway, fix permissions, auth, firewalls. Essential for vibe-coding setups.

Configure Nginx for reverse proxy, load balancing, SSL termination, and high-performance static serving.

Instinct-based learning system for OpenClaw. Analyzes sessions, detects patterns, creates atomic learnings with confidence scoring, and suggests optimizations for self-evolution. Use when: you want your AI agent to learn from its own behavior, improve over time, discover optimization opportunities, or build a self-improving automation system. Don't use when: static agent behavior is preferred.

ClawHub 安装与配置 | ClawHub setup. 帮助中文用户安装 ClawHub、配置镜像（如阿里云）、 找技能（发现/推荐）、以及技能的安装/更新/启用/禁用。 Use when: 安装 clawhub, clawhub 怎么用, 找技能/找 skill, 有什么技能可以/有什么 skill 可以, 找一个技能/找一个 skill, 搜索技能/搜索 skill, 安装技能/安装 skill, 更新技能/更新 skill, 管理技能/管理 skill, clawhub init, clawhub install, clawhub search, clawhub update, openclaw skills.