my-skill/
├── SKILL.md           # 必須：技能定義文件（YAML frontmatter + Markdown 指令）
├── _meta.json         # 可選：ClawHub 發(fā)布元數(shù)據(jù)
├── .clawhub/          # 可選：ClawHub 安裝來源信息
│   └── origin.json
├── scripts/           # 可選：可執(zhí)行腳本
├── references/        # 可選：參考文檔（按需加載，不常駐上下文）
└── assets/            # 可選：模板、圖標等靜態(tài)資源

---
name: my-skill                    # 必須：技能標識符，小寫字母+連字符
description: 技能描述...           # 必須：AI 根據(jù)這個決定何時調(diào)用該技能
version: 1.0.0                    # 可選：版本號
author: your-name                 # 可選：作者
user-invocable: true              # 可選：是否可通過斜杠命令手動調(diào)用
disable-model-invocation: false   # 可選：設(shè)為 true 則禁止 AI 自動觸發(fā)
triggers:                         # 可選：觸發(fā)關(guān)鍵詞列表
  - keyword1
  - keyword2
metadata:                         # 可選：擴展元數(shù)據(jù)
  openclaw:
    requires:
      env:
        - MY_API_KEY              # 需要的環(huán)境變量
      bins:
        - python                  # 需要的系統(tǒng)命令
---

---
name: summarize-pro
description: When user asks to summarize text, articles, documents, meetings,
  emails, YouTube transcripts, books, PDFs, reports, conversations, or any
  long content. Also handles bullet points, key takeaways, action items,
  TL;DR, ELI5, executive summaries...
---

---
name: 12306-query
description: 中國鐵路 12306 火車票余票查詢。使用官方 API 查詢車次、票價、余票信息。
  支持按出發(fā)站/到達站/日期篩選。當用戶需要查詢火車票信息（余票、車次、票價、時刻表）時使用此技能。
---

---
name: firecrawl
description: Web search and scraping via Firecrawl API. Use when you need to
  search the web, scrape websites (including JS-heavy pages), crawl entire
  sites, or extract structured data from web pages. Requires FIRECRAWL_API_KEY
  environment variable.
---

---
name: tavily-search-skill
description: Tavily API integration with real-time quota management and paid mode toggle.
version: 1.0.4
author: JayeGT002
triggers:
  - tavily
  - search
  - web search
---

---
name: agent-browser
description: Headless browser automation CLI optimized for AI agents with
  accessibility tree snapshots and ref-based element selection
metadata: {"clawdbot":{"emoji":"??","requires":{"commands":["agent-browser"]}}}
---

# Summarize Pro — Your AI Summarization Engine
You are a powerful text summarizer...
## Examples
User: "summarize this: [pastes long article]"
User: "tldr: [pastes text]"
User: "eli5: quantum computing"
## FEATURE 1: Quick Summary (Default)
When user pastes text or says "summarize this":
1. Analyze the text length and content type
2. Produce a summary in the user's default format
?? SUMMARY
━━━━━━━━━━━━━━━━━━
[3-5 bullet points capturing the main ideas]
?? Stats: [X] words → [Y] words ([Z]% reduction)

## Vetting Protocol

### Step 1: Source Check
- [ ] Where did this skill come from?
- [ ] Is the author known/reputable?

### Step 2: Code Review (MANDATORY)
?? REJECT IMMEDIATELY IF YOU SEE:
? curl/wget to unknown URLs
? Sends data to external servers
? Uses eval() or exec() with external input
...

### Step 3: Permission Scope
### Step 4: Risk Classification

#!/usr/bin/env python3
"""12306 火車票余票查詢腳本"""
# 1. 常量定義 — API 地址和緩存數(shù)據(jù)
TICKET_QUERY_URL = "https://kyfw.12306.cn/otn/leftTicket/query"
COMMON_STATIONS = {
    "北京": "BJP",
    "上海": "SHH",
    "廣州": "GZQ",
    # ...
}
# 2. 輔助函數(shù) — 站點編碼查詢
def get_station_code(station_name):
    """先查緩存，再查 API"""
    if station_name in COMMON_STATIONS:
        return COMMON_STATIONS[station_name]
    # 從 12306 官方 JS 文件解析...
# 3. 核心函數(shù) — 調(diào)用 API 查詢
def query_tickets(from_station, to_station, date):
    from_code = get_station_code(from_station)
    to_code = get_station_code(to_station)
    session = requests.Session()
    params = {
        "leftTicketDTO.train_date": date,
        "leftTicketDTO.from_station": from_code,
        "leftTicketDTO.to_station": to_code,
        "purpose_codes": "ADULT",
    }
    response = session.get(TICKET_QUERY_URL, params=params, headers=headers, timeout=10)
    # 解析返回數(shù)據(jù)...
# 4. 格式化輸出 — 表格形式，AI 友好
def format_tickets(tickets):
    print("{:<8} {:<10} {:<10} {:<8} ...".format(
        "Train", "Depart", "Arrive", "Duration", ...))
    for t in tickets:
        print(...)
# 5. 命令行入口
def main():
    if len(sys.argv) < 4:
        print("用法：python query_tickets.py <出發(fā)站> <到達站> <日期>")
        sys.exit(1)
    # ...
if __name__ == "__main__":
    main()

## 快速開始

使用腳本查詢余票：
?```bash
python scripts/query_tickets.py <出發(fā)站> <到達站> <日期>
?```

日期格式：YYYY-MM-DD（如 2026-04-01）

#!/bin/bash

# 1. 環(huán)境變量校驗 — 最重要的第一步
if [ -z "$TAVILY_API_KEY" ]; then
    echo "ERROR: TAVILY_API_KEY environment variable is not set" >&2
    exit 1
fi

# 2. 依賴檢查
if ! command -v curl &> /dev/null; then
    echo '{"error":"curl is not installed"}'
    exit 1
fi
if ! command -v jq &> /dev/null; then
    echo '{"error":"jq is not installed"}'
    exit 1
fi

# 3. 核心搜索函數(shù) — 帶重試機制
perform_search() {
    local query="$1"
    local max_results="${2:-5}"
    
    local response=$(curl -s -X POST "$TAVILY_ENDPOINT" \
        -H "Content-Type: application/json" \
        -d "{
            \"api_key\": \"$TAVILY_API_KEY\",
            \"query\": \"$query\",
            \"max_results\": $max_results
        }")
    
    # 輸出 JSON 格式結(jié)果
    echo "$body" | jq '{query: .query, results: .results, quota_info: ...}'
}

# 4. 子命令路由
case "$1" in
    --usage)        show_usage ;;
    --status)       show_status ;;
    --toggle-paid-mode) toggle_paid_mode ;;
    *)              perform_search "$1" "$2" "$3" ;;
esac

## Core Workflow
?```bash
# 1. Navigate and snapshot
agent-browser open https://example.com
agent-browser snapshot -i --json

# 2. Parse refs from JSON, then interact
agent-browser click @e2
agent-browser fill @e3 "text"

# 3. Re-snapshot after page changes
agent-browser snapshot -i --json
?```

## Key Commands
### Navigation
?```bash
agent-browser open <url>
agent-browser back | forward | reload | close
?```

### Interactions (Ref-based)
?```bash
agent-browser click @e2
agent-browser fill @e3 "text"
agent-browser hover @e4
?```