【Skill 总结】所有技能完整总结

最后更新: 2026-03-21 22:51:42

• 共有 16 个 Skill

agent-browser

原标题：Browser Automation with agent-browser

Installation

npm recommended

npm install -g agent-browser
agent-browser install
agent-browser install --with-deps

From Source

git clone https://github.com/vercel-labs/agent-browser
cd agent-browser
pnpm install
pnpm build
agent-browser install

Quick start

agent-browser open         # Navigate to page
agent-browser snapshot -i       # Get interactive elements with refs
agent-browser click @e1         # Click element by ref
agent-browser fill @e2 "text"   # Fill input by ref
agent-browser close             # Close browser

Core workflow

• Navigate: agent-browser open
• Snapshot: agent-browser snapshot -i (returns elements with refs like @e1, @e2)
• Interact using refs from the snapshot
• Re-snapshot after navigation or significant DOM changes

Commands

Navigation

agent-browser open       # Navigate to URL
agent-browser back            # Go back
agent-browser forward         # Go forward
agent-browser reload          # Reload page
agent-browser close           # Close browser

Snapshot (page analysis)

agent-browser snapshot            # Full accessibility tree
agent-browser snapshot -i         # Interactive elements only (recommended)
agent-browser snapshot -c         # Compact output
agent-browser snapshot -d 3       # Limit depth to 3
agent-browser snapshot -s "#main" # Scope to CSS selector

Interactions (use @refs from snapshot)

agent-browser click @e1           # Click
agent-browser dblclick @e1        # Double-click
agent-browser focus @e1           # Focus element
agent-browser fill @e2 "text"     # Clear and type
agent-browser type @e2 "text"     # Type without clearing
agent-browser press Enter         # Press key
agent-browser press Control+a     # Key combination
agent-browser keydown Shift       # Hold key down
agent-browser keyup Shift         # Release key
agent-browser hover @e1           # Hover
agent-browser check @e1           # Check checkbox
agent-browser uncheck @e1         # Uncheck checkbox
agent-browser select @e1 "value"  # Select dropdown
agent-browser scroll down 500     # Scroll page
agent-browser scrollintoview @e1  # Scroll element into view
agent-browser drag @e1 @e2        # Drag and drop
agent-browser upload @e1 file.pdf # Upload files

Get information

agent-browser get text @e1        # Get element text
agent-browser get html @e1        # Get innerHTML
agent-browser get value @e1       # Get input value
agent-browser get attr @e1 href   # Get attribute
agent-browser get title           # Get page title
agent-browser get url             # Get current URL
agent-browser get count ".item"   # Count matching elements
agent-browser get box @e1         # Get bounding box

Check state

agent-browser is visible @e1      # Check if visible
agent-browser is enabled @e1      # Check if enabled
agent-browser is checked @e1      # Check if checked

Screenshots & PDF

agent-browser screenshot          # Screenshot to stdout
agent-browser screenshot path.png # Save to file
agent-browser screenshot --full   # Full page
agent-browser pdf output.pdf      # Save as PDF

Video recording

agent-browser record start ./demo.webm    # Start recording (uses current URL + state)
agent-browser click @e1                   # Perform actions
agent-browser record stop                 # Stop and save video
agent-browser record restart ./take2.webm # Stop current + start new recording

Recording creates a fresh context but preserves cookies/storage from your session. If no URL is provided, it automatically returns to your current page. For smooth demos, explore first, then start recording.

Wait

agent-browser wait @e1                     # Wait for element
agent-browser wait 2000                    # Wait milliseconds
agent-browser wait --text "Success"        # Wait for text
agent-browser wait --url "/dashboard"    # Wait for URL pattern
agent-browser wait --load networkidle      # Wait for network idle
agent-browser wait --fn "window.ready"     # Wait for JS condition

Mouse control

agent-browser mouse move 100 200      # Move mouse
agent-browser mouse down left         # Press button
agent-browser mouse up left           # Release button
agent-browser mouse wheel 100         # Scroll wheel

Semantic locators (alternative to refs)

agent-browser find role button click --name "Submit"
agent-browser find text "Sign In" click
agent-browser find label "Email" fill "user@test.com"
agent-browser find first ".item" click
agent-browser find nth 2 "a" text

Browser settings

agent-browser set viewport 1920 1080      # Set viewport size
agent-browser set device "iPhone 14"      # Emulate device
agent-browser set geo 37.7749 -122.4194   # Set geolocation
agent-browser set offline on              # Toggle offline mode
agent-browser set headers '{"X-Key":"v"}' # Extra HTTP headers
agent-browser set credentials user pass   # HTTP basic auth
agent-browser set media dark              # Emulate color scheme

Cookies & Storage

agent-browser cookies                     # Get all cookies
agent-browser cookies set name value      # Set cookie
agent-browser cookies clear               # Clear cookies
agent-browser storage local               # Get all localStorage
agent-browser storage local key           # Get specific key
agent-browser storage local set k v       # Set value
agent-browser storage local clear         # Clear all

Network

agent-browser network route               # Intercept requests
agent-browser network route  --abort      # Block requests
agent-browser network route  --body '{}'  # Mock response
agent-browser network unroute [url]            # Remove routes
agent-browser network requests                 # View tracked requests
agent-browser network requests --filter api    # Filter requests

Tabs & Windows

agent-browser tab                 # List tabs
agent-browser tab new [url]       # New tab
agent-browser tab 2               # Switch to tab
agent-browser tab close           # Close tab
agent-browser window new          # New window

Frames

agent-browser frame "#iframe"     # Switch to iframe
agent-browser frame main          # Back to main frame

Dialogs

agent-browser dialog accept [text]  # Accept dialog
agent-browser dialog dismiss        # Dismiss dialog

JavaScript

agent-browser eval "document.title"   # Run JavaScript

State management

agent-browser state save auth.json    # Save session state
agent-browser state load auth.json    # Load saved state

Example: Form submission

agent-browser open https://example.com/form
agent-browser snapshot -i
原标题：Output shows: textbox "Email" [ref=e1], textbox "Password" [ref=e2], button "Submit" [ref=e3]
agent-browser fill @e1 "user@example.com"
agent-browser fill @e2 "password123"
agent-browser click @e3
agent-browser wait --load networkidle
agent-browser snapshot -i  # Check result

Example: Authentication with saved state

原标题：Login once
agent-browser open https://app.example.com/login
agent-browser snapshot -i
agent-browser fill @e1 "username"
agent-browser fill @e2 "password"
agent-browser click @e3
agent-browser wait --url "/dashboard"
agent-browser state save auth.json
原标题：Later sessions: load saved state
agent-browser state load auth.json
agent-browser open https://app.example.com/dashboard

Sessions (parallel browsers)

agent-browser --session test1 open site-a.com
agent-browser --session test2 open site-b.com
agent-browser session list

JSON output (for parsing)

Add --json for machine-readable output:

agent-browser snapshot -i --json
agent-browser get text @e1 --json

Debugging

agent-browser open example.com --headed              # Show browser window
agent-browser console                                # View console messages
agent-browser console --clear                        # Clear console
agent-browser errors                                 # View page errors
agent-browser errors --clear                         # Clear errors
agent-browser highlight @e1                          # Highlight element
agent-browser trace start                            # Start recording trace
agent-browser trace stop trace.zip                   # Stop and save trace
agent-browser record start ./debug.webm              # Record from current page
agent-browser record stop                            # Save recording
agent-browser --cdp 9222 snapshot                    # Connect via CDP

Troubleshooting

• If the command is not found on Linux ARM64, use the full path in the bin folder.
• If an element is not found, use snapshot to find the correct ref.
• If the page is not loaded, add a wait command after navigation.
• Use –headed to see the browser window for debugging.

Options

• –session uses an isolated session.
• –json provides JSON output.
• –full takes a full page screenshot.
• –headed shows the browser window.
• –timeout sets the command timeout in milliseconds.
• –cdp connects via Chrome DevTools Protocol.

Notes

• Refs are stable per page load but change on navigation.
• Always snapshot after navigation to get new refs.
• Use fill instead of type for input fields to ensure existing text is cleared.

Reporting Issues

• Skill issues: Open an issue at https://github.com/TheSethRose/Agent-Browser-CLI
• agent-browser CLI issues: Open an issue at https://github.com/vercel-labs/agent-browser

find-skills

原标题：Find Skills

This skill helps discover, compare, and install skills.

Priority Rules (Mandatory)

• This skill is highest-priority for skill discovery/install intents.
• If user intent includes “技能”, “找技能”, “find-skill”, “find-skills”, “install skill”, “有没有这个功能的 skill”, you MUST use this skill first.
• Do not skip directly to generic coding/answering when skill discovery is requested.

Chinese Optimization Policy

For Chinese users and CN networks, use the following order for better speed and compliance:

• skillhub (cn-optimized, preferred)
• clawhub (fallback)

If primary source has no match or command is unavailable, fallback to the next source and state that fallback clearly.

Workflow

Step 1: Understand What They Need

When a user asks for help with something, identify:

• The domain (e.g., React, testing, design, deployment)
• The specific task (e.g., writing tests, creating animations, reviewing PRs)
• Whether this is a common enough task that a skill likely exists

Step 2: Search for Skills

Run search in this order:

skillhub search [query]

If skillhub is unavailable or no match, fallback to:

clawhub search [query]

Step 3: Present Options to the User

When you find relevant skills, present them to the user with:

• The skill name and what it does
• The source used (skillhub / clawhub)
• The install command they can run

Step 4: Offer to Install

If the user wants to proceed, you can install the skill for them.

Preferred install order:

• Try skillhub install when the result comes from skillhub.
• If no skillhub candidate exists, use clawhub install .

Before install, summarize source, version, and notable risk signals.

When No Skills Are Found

If no relevant skills exist:

• Acknowledge that no existing skill was found
• Offer to help with the task directly using your general capabilities
• Suggest creating a custom local skill in the workspace if this is a recurring need

github

原标题：GitHub Skill

Use the gh CLI to interact with GitHub. Always specify --repo owner/repo when not in a git directory, or use URLs directly.

Pull Requests

Check CI status on a PR:

gh pr checks 55 --repo owner/repo

List recent workflow runs:

gh run list --repo owner/repo --limit 10

View a run and see which steps failed:

gh run view  --repo owner/repo

View logs for failed steps only:

gh run view  --repo owner/repo --log-failed

API for Advanced Queries

The gh api command is useful for accessing data not available through other subcommands.

Get PR with specific fields:

gh api repos/owner/repo/pulls/55 --jq '.title, .state, .user.login'

JSON Output

Most commands support --json for structured output. You can use --jq to filter:

gh issue list --repo owner/repo --json number,title --jq '.[] | "\(.number): \(.title)"'

humanize-chinese

原标题：Humanize Chinese AI Text

Comprehensive CLI for detecting and transforming Chinese AI-generated text. Makes robotic AI writing natural and human-like with 6 specialized writing style transforms.

NEW in v1.1: Style transforms (知乎/小红书/公众号/口语化/学术/文艺), enhanced detection (16 patterns), emotional analysis

Quick Start

原标题：Detect AI patterns (16 categories)
python scripts/detect_cn.py text.txt
原标题：Humanize text
python scripts/humanize_cn.py text.txt -o clean.txt
原标题：Scene-specific humanization
python scripts/humanize_cn.py text.txt --scene social  # Social media
python scripts/humanize_cn.py text.txt --scene tech    # Tech blog
python scripts/humanize_cn.py text.txt --scene formal  # Formal article
原标题：NEW: Apply writing styles
python scripts/humanize_cn.py text.txt --style zhihu -o zhihu.txt
python scripts/humanize_cn.py text.txt --style xiaohongshu -o xhs.txt
python scripts/style_cn.py text.txt --style casual -o casual.txt
原标题：List all available styles
python scripts/style_cn.py --list
原标题：Compare before/after
python scripts/compare_cn.py text.txt -o clean.txt

Detection Categories

The analyzer checks for 16 pattern categories specific to Chinese AI text (v1.1 added 4 new patterns):

Critical (Immediate AI Detection)

High Signal

Medium Signal

Style Signal

Writing Style Transforms (NEW in v1.1)

Transform text into 6 specialized Chinese writing styles:

| Style | Name | Description | Best For | |——-|——|————-|———-| | casual | 口语化风格 | Like chatting with friends — natural, relaxed | Social media, messaging | | zhihu | 知乎风格 | Rational, in-depth, with personal opinions | Q&A platforms, thoughtful analysis | | xiaohongshu | 小红书风格 | Enthusiastic, emoji-rich, product-focused | Lifestyle sharing, reviews, recommendations | | wechat | 公众号风格 | Storytelling, engaging, relatable | WeChat articles, newsletters | | academic | 学术风格 | Rigorous but not stiff, precise terminology | Academic papers, research reports | | literary | 文艺风格 | Poetic, imagery-rich, metaphorical | Creative writing, essays |

Usage

原标题：Apply style directly
python scripts/style_cn.py input.txt --style zhihu -o output.txt
原标题：Combine humanization + style
python scripts/humanizecn.py aitext.txt --style xiaohongshu -o natural.txt
原标题：List all styles
python scripts/style_cn.py --list

Style Features

Casual (口语化)

• Removes formal structure (首先/其次/最后)
• Adds colloquial connectors (说实话, 确实, 其实)
• Includes tone particles (吧, 呢, 啊)
• Light emoji usage

Zhihu (知乎)

• Personal opinion markers (从我的经验来看, 个人认为)
• Data/evidence support (实测发现, 根据XX调研)
• Example-driven (举个例子)
• Logical but conversational

Xiaohongshu (小红书)

• Enthusiastic openers (姐妹们！分享一下～)
• High emoji density (😊✨💯)
• Intensifiers (超级, 巨, 绝绝子, yyds)
• Hashtags (#好物分享)
• Short paragraphs

Wechat (公众号)

• Story-driven openings
• Questions for engagement (你有没有想过)
• Relatable scenarios
• Emotional connection

Academic (学术)

• Removes colloquialisms
• Precise terminology
• Formal connectors (研究表明, 数据显示)
• Reduces emotional expressions

Literary (文艺)

• Metaphors and imagery
• Poetic language
• Descriptive phrases (在XX的光影里)
• Artistic expressions

Scripts

detect_cn.py — Scan Chinese AI Patterns

python scripts/detect_cn.py essay.txt
python scripts/detect_cn.py essay.txt -j  # JSON output
python scripts/detect_cn.py essay.txt -s  # score only
echo "文本" | python scripts/detect_cn.py

Output:

• AI feature statistics (by category)
• AI probability score (low/medium/high/very high)
• Auto-fixable patterns marked
• Perplexity and burstiness indicators

humanize_cn.py — Transform to Human-Like

python scripts/humanize_cn.py essay.txt
python scripts/humanize_cn.py essay.txt -o output.txt
python scripts/humanize_cn.py essay.txt --scene social  # Social media style
python scripts/humanize_cn.py essay.txt -a              # Aggressive mode

Scene Parameters (–scene):

• social: Social media (casual, conversational)
• tech: Tech blog (professional but approachable)
• formal: Formal article (rigorous but natural)
• chat: Chat/dialogue (friendly, concise)

Auto-fixes:

• Remove three-part structure (首先/其次/最后)
• Replace mechanical connectors (值得注意的是 → 注意/要提醒的是)
• Simplify empty words (赋能 → 帮助/提升, 闭环 → 完整流程)
• Reduce punctuation density (em dash, semicolon)
• Control rhetoric frequency (对偶, 排比, 比喻)

Aggressive Mode (-a):

• Add colloquial expressions
• Inject emotional color
• Vary sentence rhythm
• Add personal perspective

compare_cn.py — Before/After Analysis

python scripts/compare_cn.py essay.txt
python scripts/compare_cn.py essay.txt --scene tech -o clean.txt

Shows AI feature comparison and score changes before/after transformation.

style_cn.py — Writing Style Transform (NEW)

python scripts/stylecn.py essay.txt --style zhihu -o essayzhihu.txt
python scripts/stylecn.py blog.txt --style xiaohongshu -o blogxhs.txt
python scripts/style_cn.py --list  # Show all available styles

Supported styles: casual, zhihu, xiaohongshu, wechat, academic, literary

Transform text into specific Chinese writing styles with style-appropriate vocabulary, tone, and structure.

Workflow

• Scan for detection risk:

   python scripts/detect_cn.py document.txt
   ``

Transform with comparison:

   `bash
   python scripts/comparecn.py document.txt --scene tech -o documentv2.txt
   `

Verify improvement:

   `bash
   python scripts/detectcn.py documentv2.txt -s
   `

Manual review for content quality and scene appropriateness
AI Probability Scoring
| Rating | Criteria |
|--------|----------|
| Very High | Three-part structure, mechanical connectors, or empty grand words present |
| High | >20 issues OR issue density >3% |
| Medium | >10 issues OR issue density >1.5% |
| Low | <10 issues AND density <1.5% |
Scene-Specific Guidelines
Social Media (社交媒体)
Style: Casual, conversational, like chatting with friends

✅ Short paragraphs (1-3 sentences)
✅ Colloquial expressions (说实话, 没想到, 真的绝了)
✅ Specific details (product names, locations, personal feelings)
✅ Emoji and hashtags
❌ Avoid: 值得注意的是, 总而言之
❌ Avoid: Long paragraphs, complex sentences
Tech Blog (技术博客)
Style: Professional but approachable, can be humorous

✅ Specific tech stack, tool names
✅ Code examples, performance data
✅ Real experiences ("踩过的坑", "实测效果")
✅ Clear structure with headings (not numbered lists)
❌ Avoid: 赋能, 闭环, 生态
❌ Avoid: 首先/其次/最后structure
Formal Article (正式文章)
Style: Objective, rigorous, but natural

✅ Clear logic with proper evidence
✅ Precise academic expressions
✅ Cited research sources
✅ Data and charts supporting arguments
❌ Avoid: Excessive rhetoric (对偶, 排比)
❌ Avoid: Empty grand words
Chat/Dialogue (对话场景)
Style: Friendly, patient, genuine

✅ Concise, targeted responses
✅ Empathy and understanding
✅ Direct solutions
✅ Moderate emoji use
❌ Avoid: 很高兴为您服务 (template phrases)
❌ Avoid: Lengthy explanations, repetitive questions
Customizing Patterns
Edit scripts/patterns_cn.json to add/modify:

aivocabularycn — Chinese AI high-frequency words

fillerphrasescn — Clichés and replacements

emptywordscn — Empty grand vocabulary

rhetoric_limits — Rhetoric frequency limits

scene_styles — Scene-specific style configs

Batch Processing

Reference
Based on comprehensive Chinese AI writing research:

Tencent News: "Deconstructing 'AI Flavor': Why We Dislike AI Writing"
53AI: "Detection and Optimization of Article 'AI Flavor'"
AIGCleaner and other Chinese de-AI tools
Wikipedia: "Signs of AI Writing" (English reference)
Key insights:

Perplexity: AI text has low perplexity (predictable word choices)
Burstiness: AI text has low burstiness (uniform sentence structure)
Emotion: AI text lacks strong opinions and personal color
multi-search-engine

name: "multi-search-engine"
description: "Multi search engine integration with 17 engines (8 CN + 9 Global). Supports advanced search operators, time filters, site search, privacy engines, and WolframAlpha knowledge queries. No API keys required."
原标题：Multi Search Engine v2.0.1
Integration of 17 search engines for web crawling without API keys.
Search Engines
Domestic (8)

Baidu: https://www.baidu.com/s?wd={keyword}

Bing CN: https://cn.bing.com/search?q={keyword}&ensearch=0

Bing INT: https://cn.bing.com/search?q={keyword}&ensearch=1

360: https://www.so.com/s?q={keyword}

Sogou: https://sogou.com/web?query={keyword}

WeChat: https://wx.sogou.com/weixin?type=2&query={keyword}

Toutiao: https://so.toutiao.com/search?keyword={keyword}

Jisilu: https://www.jisilu.cn/explore/?keyword={keyword}

International (9)


Google: https://www.google.com/search?q={keyword}

Google HK: https://www.google.com.hk/search?q={keyword}

DuckDuckGo: https://duckduckgo.com/html/?q={keyword}

Yahoo: https://search.yahoo.com/search?p={keyword}

Startpage: https://www.startpage.com/sp/search?query={keyword}

Brave: https://search.brave.com/search?q={keyword}

Ecosia: https://www.ecosia.org/search?q={keyword}

Qwant: https://www.qwant.com/?q={keyword}

WolframAlpha: https://www.wolframalpha.com/input?i={keyword}

Quick Examples

Advanced Operators
| Operator | Example | Description |
|----------|---------|-------------|
| site: | site:github.com python | Search within site |
| filetype: | filetype:pdf report | Specific file type |
| "" | "machine learning" | Exact match |
| - | python -snake | Exclude term |
| OR | cat OR dog | Either term |
Time Filters
| Parameter | Description |
|-----------|-------------|
| tbs=qdr:h | Past hour |
| tbs=qdr:d | Past day |
| tbs=qdr:w | Past week |
| tbs=qdr:m | Past month |
| tbs=qdr:y | Past year |
Privacy Engines

DuckDuckGo: No tracking
Startpage: Google results + privacy
Brave: Independent index
Qwant: EU GDPR compliant
Bangs Shortcuts (DuckDuckGo)
| Bang | Destination |
|------|-------------|
| !g | Google |
| !gh | GitHub |
| !so | Stack Overflow |
| !w | Wikipedia |
| !yt | YouTube |
WolframAlpha Queries

Math: integrate x^2 dx

Conversion: 100 USD to CNY

Stocks: AAPL stock

Weather: weather in Beijing

Documentation

references/advanced-search.md - Domestic search guide

references/international-search.md - International search guide

CHANGELOG.md - Version history

License
MIT
nano-pdf

name: nano-pdf
description: Edit PDFs with natural-language instructions using the nano-pdf CLI.
homepage: https://pypi.org/project/nano-pdf/
metadata: {"clawdbot":{"emoji":"📄","requires":{"bins":["nano-pdf"]},"install":[{"id":"uv","kind":"uv","package":"nano-pdf","bins":["nano-pdf"],"label":"Install nano-pdf (uv)"}]}}
原标题：nano-pdf
Use nano-pdf to apply edits to a specific page in a PDF using a natural-language instruction.
Quick start

Notes:

Page numbers are 0-based or 1-based depending on the tool’s version/config; if the result looks off by one, retry with the other.
Always sanity-check the output PDF before sending it out.
obsidian

name: obsidian
description: Work with Obsidian vaults (plain Markdown notes) and automate via obsidian-cli.
homepage: https://help.obsidian.md
metadata: {"clawdbot":{"emoji":"💎","requires":{"bins":["obsidian-cli"]},"install":[{"id":"brew","kind":"brew","formula":"yakitrak/yakitrak/obsidian-cli","bins":["obsidian-cli"],"label":"Install obsidian-cli (brew)"}]}}
原标题：Obsidian
Obsidian vault = a normal folder on disk.
Vault structure (typical)

Notes: *.md (plain text Markdown; edit with any editor)

Config: .obsidian/ (workspace + plugin settings; usually don’t touch from scripts)

Canvases: *.canvas (JSON)

Attachments: whatever folder you chose in Obsidian settings (images/PDFs/etc.)
Find the active vault(s)
Obsidian desktop tracks vaults here (source of truth):

~/Library/Application Support/obsidian/obsidian.json

obsidian-cli resolves vaults from that file; vault name is typically the folder name (path suffix).
Fast “what vault is active / where are the notes?”

If you’ve already set a default: obsidian-cli print-default --path-only

Otherwise, read ~/Library/Application Support/obsidian/obsidian.json and use the vault entry with "open": true.

Notes

Multiple vaults common (iCloud vs ~/Documents, work/personal, etc.). Don’t guess; read config.

Avoid writing hardcoded vault paths into scripts; prefer reading the config or using print-default.

obsidian-cli quick start
Pick a default vault (once):

obsidian-cli set-default ""

obsidian-cli print-default / obsidian-cli print-default --path-only

Search

obsidian-cli search "query" (note names)

obsidian-cli search-content "query" (inside notes; shows snippets + lines)

Create

obsidian-cli create "Folder/New note" --content "..." --open

Requires Obsidian URI handler (obsidian://…) working (Obsidian installed).

Avoid creating notes under “hidden” dot-folders (e.g. .something/...) via URI; Obsidian may refuse.

Move/rename (safe refactor)

obsidian-cli move "old/path/note" "new/path/note"

Updates [[wikilinks]] and common Markdown links across the vault (this is the main win vs mv).

Delete

obsidian-cli delete "path/note"

Prefer direct edits when appropriate: open the .md file and change it; Obsidian will pick it up.
pdf-generator

name: pdf-generator
description: 使用Python生成符合中国公文规范的中文PDF文档，先生成docx再转PDF，支持方正小标宋、楷体、仿宋等公文字体，自动处理页边距、行间距、页码、落款等格式要求。使用当用户需要将信息整理成标准公文格式的PDF时。
原标题：PDF Generator - 中文公文PDF生成工具
快速开始
通过先生成docx再转PDF的方式，生成符合中国公文规范的PDF文档。
系统环境
Python库

python-docx (docx生成)
libreoffice (docx → PDF转换)
可用中文字体

方正小标宋简体 (/root/.local/share/fonts/FZXiaoBiaoSong-Regular.ttf)
楷体_GB2312 (/root/.local/share/fonts/KaiTi-GB2312.ttf)
仿宋_GB2312 (/root/.local/share/fonts/FangSong-GB2312.ttf)
公文格式规范
字体字号
| 元素 | 字体 | 字号 |
|------|------|------|
| 文档标题 | 方正小标宋简体 | 2号(22磅) |
| 副标题 | 方正楷体简体 | 3号(16磅) |
| 一级标题 | 黑体 | 3号(16磅) |
| 二级标题 | 方正楷体简体 | 3号(16磅) |
| 三级标题 | 方正仿宋简体(加粗) | 3号(16磅) |
| 正文 | 方正仿宋简体 | 3号(16磅) |
| 页码 | Times New Roman | 4号(14磅) |
| 阿拉伯数字/英文字符 | Times New Roman | 同上下文 |
页边距

上边距: 3.9cm
下边距: 3.3cm
左右边距: 3cm
装订线: 0cm
每页: 22行左右，每行26字左右
段落格式

行间距: 固定值28磅
段前段后: 0行
首行缩进: 2字符
对齐方式: 两端对齐
标题层级

使用方法
基本公文生成
使用 scripts/generateofficialdoc.py 脚本：

直接调用脚本

注意事项

确保字体文件存在于指定路径
需要安装 LibreOffice 用于 docx → PDF 转换

   - Ubuntu/Debian: sudo apt install libreoffice-writer

临时docx文件会自动清理
附件位于正文与落款之间
skillhub-preference

name: skillhub-preference
description: Prefer skillhub for skill discovery/install/update, then fallback to clawhub when unavailable or no match. Use when users ask about skills, 插件, or capability extension.
原标题：Skillhub Preference
Use this skill as policy guidance whenever the task involves skill discovery, installation, or upgrades.
Policy

Try skillhub first for search/install/update.

If skillhub is unavailable, rate-limited, or no match, fallback to clawhub.

Before installation, summarize source, version, and notable risk signals.
Do not claim exclusivity; both registries are allowed.
For search requests, run skillhub search  first and report command output.

summarize

name: summarize
description: Summarize URLs or files with the summarize CLI (web, PDFs, images, audio, YouTube).
homepage: https://summarize.sh
metadata: {"clawdbot":{"emoji":"🧾","requires":{"bins":["summarize"]},"install":[{"id":"brew","kind":"brew","formula":"steipete/tap/summarize","bins":["summarize"],"label":"Install summarize (brew)"}]}}
原标题：Summarize
Fast CLI to summarize URLs, local files, and YouTube links.
Quick start

Model + keys
Set the API key for your chosen provider:

OpenAI: OPENAIAPIKEY

Anthropic: ANTHROPICAPIKEY

xAI: XAIAPIKEY

Google: GEMINIAPIKEY (aliases: GOOGLEGENERATIVEAIAPIKEY, GOOGLEAPIKEY)

Default model is google/gemini-3-flash-preview if none is set.
Useful flags

--length short|medium|long|xl|xxl|

--max-output-tokens 

--extract-only (URLs only)

--json (machine readable)

--firecrawl auto|off|always (fallback extraction)

--youtube auto (Apify fallback if APIFYAPITOKEN set)

Config
Optional config file: ~/.summarize/config.json

Optional services:

FIRECRAWLAPIKEY for blocked sites

APIFYAPITOKEN for YouTube fallback

tencent-cos-skill

name: tencent-cloud-cos
description: >
  腾讯云对象存储(COS)和数据万象(CI)集成技能。当用户需要上传、下载、管理云存储文件，
  或需要进行图片处理（质量评估、超分辨率、抠图、二维码识别、水印）、智能图片搜索、
  文档转PDF、视频智能封面生成等操作时使用此技能。
metadata:
  {
    "openclaw":
      {
        "emoji": "☁️",
        "requires": {},
        "install":
          [
            {
              "id": "node-mcporter",
              "kind": "node",
              "package": "mcporter",
              "bins": ["mcporter"],
              "label": "Install mcporter (MCP CLI)",
            },
            {
              "id": "node-cos-mcp",
              "kind": "node",
              "package": "cos-mcp",
              "bins": ["cos-mcp"],
              "label": "Install cos-mcp (COS MCP Server)",
            },
            {
              "id": "node-cos-sdk",
              "kind": "node",
              "package": "cos-nodejs-sdk-v5",
              "label": "Install COS Node.js SDK",
            },
          ],
      },
  }
原标题：腾讯云 COS 技能
通过 cos-mcp MCP 工具 + Node.js SDK 脚本 + COSCMD 管理腾讯云对象存储和数据万象。
首次使用 — 自动设置
当用户首次要求操作 COS 时，按以下流程操作：
步骤 1：检查当前状态

如果输出显示一切 OK（cos-mcp 已安装、凭证已配置），跳到「执行策略」。
步骤 2：如果未配置，引导用户提供凭证
告诉用户：
> 我需要你的腾讯云凭证来连接 COS 存储服务。请提供：
> 1. SecretId — 腾讯云 API 密钥 ID
> 2. SecretKey — 腾讯云 API 密钥 Key
> 3. Region — 存储桶区域（如 ap-guangzhou）
> 4. Bucket — 存储桶名称（格式 name-appid，如 mybucket-1250000000）
> 5. DatasetName（可选） — 数据万象数据集名称（仅智能搜索需要）
> 6. Domain（可选） — 自定义域名，用于替换默认的 COS 访问域名（如 cdn.example.com）
> 7. ServiceDomain（可选） — 自定义服务域名，用于自定义 COS API 请求域名
> 8. Protocol（可选） — 协议，如 https 或 http
>
> 你可以在 腾讯云控制台 > 访问管理 > API密钥管理 获取密钥，
> 在 COS 控制台 查看存储桶信息。
步骤 3：用户提供凭证后，运行自动设置

如有 DatasetName：

如需自定义域名（可选参数按需添加）：

脚本会自动：

检查并安装 mcporter（MCP 命令行工具）
检查并安装 cos-mcp 和 cos-nodejs-sdk-v5
创建/更新 ~/.mcporter/mcporter.json，写入 cos-mcp 服务器配置

将凭证写入 shell 配置文件（~/.zshrc 或 ~/.bashrc），重启后仍可用

配置 coscmd（如有 Python 环境）
验证 COS 连接
设置完成后即可开始使用。
执行策略
三种方式按优先级降级，确保操作始终可完成：

方式一：cos-mcp MCP 工具（优先） — 功能最全，支持存储 + 图片处理 + 智能搜索 + 文档媒体处理
方式二：Node.js SDK 脚本 — 通过 scripts/cos_node.mjs 执行存储操作

方式三：COSCMD 命令行 — 通过 shell 命令执行存储操作

判断方式一(mcporter)：which mcporter 且 cat ~/.mcporter/mcporter.json | grep cos-mcp 有输出。
判断方式一(直接)：尝试调用 getCosConfig MCP 工具，若返回结果则可用。
判断方式二：node -e "require('cos-nodejs-sdk-v5')" 成功则可用。
判断方式三：which coscmd 有输出则可用。
方式一：cos-mcp MCP 工具（优先）
> GitHub: https://github.com/Tencent/cos-mcp
MCP 配置模板见 references/config_template.json。
调用格式
通过 mcporter 命令行调用 cos-mcp MCP 工具：

列出所有可用工具：

判断 mcporter 是否可用：which mcporter 且 ~/.mcporter/mcporter.json 包含 cos-mcp 配置。
如果 mcporter 不可用，可回退到客户端直接调用 MCP 工具（getCosConfig 等）。
工具总览
| 类别 | 说明 |
|------|------|
| 存储操作 | 上传、下载、列出、获取签名URL |
| 图片处理 | 质量评估、超分辨率、抠图、二维码识别、水印 |
| 智能搜索 | 以图搜图、文本搜图（需预建数据集） |
| 文档媒体 | 文档转PDF、视频智能封面（异步任务） |
常用操作
> 以下示例同时展示两种调用格式。mcporter 格式省略公共前缀 mcporter call cos-mcp. 和 --config ~/.mcporter/mcporter.json --output json。
> 完整 mcporter 命令：mcporter call cos-mcp. --config ~/.mcporter/mcporter.json --output json --args ''
存储

图片处理

智能搜索（需预建数据集）

文档与媒体处理（异步任务）

工具详细参数定义见 references/api_reference.md。
方式二：Node.js SDK 脚本
> 官方文档: https://www.tencentcloud.com/zh/document/product/436/8629
当 cos-mcp 不可用时，通过 scripts/cos_node.mjs 执行存储操作。凭证从环境变量读取。
支持的环境变量：

TENCENTCOSSECRETID / TENCENTCOSSECRETKEY / TENCENTCOSREGION / TENCENTCOSBUCKET（必需）

TENCENTCOSDOMAIN / TENCENTCOSSERVICEDOMAIN / TENCENTCOS_PROTOCOL（可选，自定义域名）

常用命令
> 以下省略 node {baseDir}/scripts/cosnode.mjs 前缀。完整格式：node {baseDir}/scripts/cosnode.mjs  [options]

所有命令输出 JSON 格式，success: true 表示成功，退出码 0。
限制
仅支持存储操作，不支持图片处理、智能搜索、文档转换。
方式三：COSCMD 命令行
> 官方文档: https://www.tencentcloud.com/zh/document/product/436/10976
当方式一和方式二均不可用时使用。配置持久化在 ~/.cos.conf。
自定义域名支持（有限）：

ServiceDomain — 对应 coscmd 的 -e ENDPOINT 参数，设置后 Region 失效

Protocol — 若为 http，对应 coscmd 的 --do-not-use-ssl 参数

Domain — COSCMD 不支持 CDN 自定义域名
常用命令

限制
仅支持存储操作，不支持图片处理、智能搜索、文档转换。
功能对照表
| 功能 | 方式一 cos-mcp | 方式二 Node SDK | 方式三 COSCMD |
|------|:-:|:-:|:-:|
| 上传文件 | ✅ | ✅ | ✅ |
| 上传字符串/Base64 | ✅ | ✅ | ❌ |
| 通过 URL 上传 | ✅ | ❌ | ❌ |
| 下载文件 | ✅ | ✅ | ✅ |
| 列出文件 | ✅ | ✅ | ✅ |
| 获取签名 URL | ✅ | ✅ | ✅ |
| 删除文件 | ❌ | ✅ | ✅ |
| 查看文件信息 | ❌ | ✅ | ✅ |
| 递归上传/下载目录 | ❌ | ❌ | ✅ |
| 图片处理（CI） | ✅ | ❌ | ❌ |
| 智能搜索 | ✅ | ❌ | ❌ |
| 文档转 PDF | ✅ | ❌ | ❌ |
| 视频智能封面 | ✅ | ❌ | ❌ |
使用规范

首次使用先运行 {baseDir}/scripts/setup.sh --check-only 检查环境

mcporter 调用必须带 --config ~/.mcporter/mcporter.json 和 --output json

凭证不明文展示：引导用户自行通过 setup.sh 或编辑配置文件设置
所有文件路径（objectKey/cospath/--key）为存储桶内的相对路径，如 images/photo.jpg

图片处理/智能搜索/文档转换仅方式一可用，不可用时明确告知用户
异步任务（文档转换、视频封面）需通过 jobId 轮询结果

上传后主动获取链接：上传完成后调用 getObjectUrl 或 sign-url 返回访问链接

错误处理：调用失败时先用 setup.sh --check-only 诊断环境问题

方式二脚本源码见 scripts/cos_node.mjs

MCP 工具详细参数见 references/api_reference.md

MCP 配置模板见 references/config_template.json

tencent-docs

name: tencent-docs
description: 腾讯文档，提供完整的腾讯文档操作能力。当用户需要操作腾讯文档时使用此skill，包括：(1) 创建各类在线文档（智能文档、Word、Excel、幻灯片、思维导图、流程图）(2) 查询、搜索文档空间与文件 (3) 管理空间节点、文件夹结构 (4) 读取文档内容 (5) 编辑操作智能表 （6）编辑操作智能文档。
homepage: https://docs.qq.com/home
metadata: {"openclaw":{"primaryEnv":"TENCENTDOCSTOKEN","category":"tencent","tencentTokenMode":"custom","tokenUrl":"https://docs.qq.com/open/document/mcp/get-token/","emoji":"📝"}}
原标题：腾讯文档 MCP 使用指南
腾讯文档 MCP 提供了一套完整的在线文档操作工具，支持创建、查询、编辑多种类型的在线文档。
支持的文档类型
| 类型 | doc_type | 推荐度 | 说明 |
|------|----------|--------|------|
| 智能文档 | smartcanvas | ⭐⭐⭐ 首选 | 排版美观，支持丰富组件 |
| Excel | excel | ⭐⭐⭐ | 数据表格专用 |
| 幻灯片 | slide | ⭐⭐⭐ | 演示文稿专用 |
| 思维导图 | mind | ⭐⭐⭐ | 知识图谱专用 |
| 流程图 | flowchart | ⭐⭐⭐ | 流程展示专用 |
| Word | word | ⭐⭐ | 传统格式，排版一般 |
| 收集表 | form | ⭐⭐ | 表单收集 |
| 智能表格 | smartsheet | ⭐⭐⭐ | 高级结构化表格，支持多视图、字段管理 |
| 白板 | board | ⭐⭐ | 在线白板 |
文档类型选择决策树

API详细参考文档
首先需要阅读文件 references/api_references.md 查看所有工具的完整API说明，该文件包含所有工具的完整调用示例、参数说明、返回值说明及API结构、枚举值说明
🎯 场景化文档指引

根据您的具体任务场景，选择相应的参考文档进行查阅：
| 任务场景 | 推荐参考文档 | 适用工具类型 | 主要用途 |
|---------|-------------|-------------|----------|
| 智能表格操作 - 处理数据表格、字段管理、记录操作 | 阅读文件 references/smartsheet_references.md | smartsheet.* 系列工具 | 智能表格专项参考，包含字段类型枚举、字段值格式参考、典型工作流示例 |
| 智能文档编辑 - 操作页面、文本、标题、待办事项等元素 | 阅读文件 references/smartcanvas_references.md | smartcanvas.* 系列工具 | 智能文档专项参考，包含元素类型说明、富文本格式枚举、典型工作流示例 |
| 在线表格操作 - 查询表格信息、获取范围数据、批量更新单元格 | 阅读文件 references/sheet_references.md | sheet.* 系列工具 | 在线表格专项参考，包含表格信息查询、范围数据获取、批量更新操作等 |
⚙️ 快速配置

在 OpenClaw 中使用时，需要先完成本地安装和注册。
安装步骤：

运行 setup.sh 完成 MCP 服务注册：

> setup.sh 会自动将腾讯文档 MCP 服务注册到 mcporter，并验证配置是否成功。
> 如果未执行 setup，所有工具调用将无法找到 tencent-docs 服务。
> ⚠️ 如果 TENCENTDOCSTOKEN 为空或未配置，请先访问 https://docs.qq.com/open/auth/mcp.html 获取 Token，并配置环境变量：export TENCENTDOCSTOKEN="你的Token值"，否则所有工具调用将返回鉴权失败。

验证安装是否成功：

🔧 调用方式
获取完整的工具列表
使用以下命令：

获取到腾讯文档 mcp服务的功能列表，然后再选择合适的工具，封装好参数，进行调用。
工具列表示例
| 工具名称 | 功能说明 | 需要阅读的参考文档 |
|---------|---------|----------|
| createsmartcanvasbymarkdown | ⭐ 创建智能文档（首选） | references/apireferences.md |
| createexcelbymarkdown | 创建 Excel 表格 | references/apireferences.md |
| createslidebymarkdown | 创建幻灯片 | references/apireferences.md |
| createmindbymarkdown | 创建思维导图 | references/apireferences.md |
| createflowchartbymermaid | 创建流程图 | references/apireferences.md |
| createwordbymarkdown | 创建 Word 文档 | references/apireferences.md |
| queryspacenode | 查询空间节点 | references/api_references.md |
| createspacenode | 创建空间节点 | references/api_references.md |
| deletespacenode | 删除空间节点 | references/api_references.md |
| searchspacefile | 搜索空间文件 | references/api_references.md |
| getcontent | 获取文档内容 | references/apireferences.md |
| batchupdatesheetrange | 批量更新表格 | references/apireferences.md |
| upload_image | 上传图片，获取 imageID 供智能表格图片字段使用 |  |
| scrapeurl | 网页剪藏：抓取网页内容并自动保存为智能文档，返回taskid用于进度查询 | references/api_references.md |
| scrapeprogress | 查询网页剪藏任务进度，与scrapeurl配合使用 | references/api_references.md |
| sheet.* | 在线表格操作（查询信息、获取范围、批量更新） | references/sheet_references.md |
| smartcanvas.* | 智能文档元素操作（页面/文本/标题/待办事项） | references/smartcanvas_references.md |
| smartsheet.* | 智能表格操作（工作表/视图/字段/记录） | references/smartsheet_references.md |
调用示例
详细调用示例和参数请阅读这个文件：references/api_references.md
获取正文内容 get_content

创建智能文档 createsmartcanvasby_markdown

创建智能表格

查询智能表中的工作表 

智能表中添加字段

mcporter call "tencent-docs.createexcelby_markdown" --args '{"title": "我的日程表", "markdown": "| 日期 | 时间 | 事项 | 地点 | 状态 | 备注 |\n|------|------|------|------|------|------|\n| 2024-03-11 | 09:00-10:00 | 团队会议 | 会议室A | 待办 | 准备项目汇报 |\n| 2024-03-11 | - | 项目文档编写 | 远程 | 进行中 | 完成需求文档 |\n| 2024-03-11 | 14:00-15:30 | 客户沟通 | 线上会议 | 已安排 | 准备演示材料 |\n| 2024-03-12 | 10:00-12:00 | 产品评审 | 会议室B | 待办 | 检查产品原型 |\n| 2024-03-12 | 15:00-16:00 | 培训学习 | 培训室 | 已安排 | AI工具使用 |\n| 2024-03-13 | 全天 | 项目开发 | 办公室 | 进行中 | 功能模块开发 |\n| 2024-03-14 | 09:30-11:00 | 周会总结 | 会议室A | 待办 | 整理本周工作 |\n| 2024-03-15 | 13:00-17:00 | 项目演示 | 客户现场 | 已安排 | 最终演示准备 |"}'

首先阅读 references目录下的所有参考文件，理解每个工具的功能和参数

创建通用文档（推荐方式）

📖 参考文档： references/apireferences.md - createsmartcanvasbymarkdown

1. 优先调用 createsmartcanvasby_markdown 创建智能文档

从返回结果中获取 file_id 和 url

📖 参考文档： references/smartcanvas_references.md - 典型工作流示例

1. 调用 smartcanvas.gettoplevel_pages 获取文档页面结构

按需调用 smartcanvas.* 工具进行增删改查：

   - 追加内容：smartcanvas.appendinsertsmartcanvasbymarkdown（Markdown 方式）
   - 新增元素：smartcanvas.createsmartcanvaselement
   - 查询元素：smartcanvas.getelementinfo / smartcanvas.getpageinfo
   - 修改元素：smartcanvas.update_element
   - 删除元素：smartcanvas.delete_element

组织文档到指定目录

📖 参考文档： references/apireferences.md - queryspacenode, createspace_node

• 调用 queryspacenode 查找目标文件夹
• 调用 createspacenode 在目标位置创建文档节点（doc_type 优先选择 smartcanvas）

搜索并读取文档

📖 参考文档： references/apireferences.md - searchspacefile, getcontent

• 调用 searchspacefile 搜索文档
• 从结果中获取 nodeid（即 fileid）
• 调用 get_content 获取文档内容

智能表格操作工作流

📖 参考文档： references/smartsheet_references.md - 典型工作流示例

从零搭建任务管理表

1. 获取工作表列表 → smartsheet.listtables（获取 sheetid）

添加字段（列）→ smartsheet.add_fields（任务名称、优先级、截止日期等）
批量写入数据 → smartsheet.add_records
（可选）创建看板视图 → smartsheet.addview（viewtype=2）
（可选）删除字段（列） → smartsheet.delete_fields

1. 获取工作表 → smartsheet.list_tables

查询记录   → smartsheet.listrecords（获取 recordid）
更新记录   → smartsheet.updaterecords（传入 recordid 和新字段值）

> 📖 更多智能表格工作流示例请参考：references/smartsheet_references.md - 典型工作流示例

注意事项

• 默认使用 smartcanvas：除非用户明确指定其他格式，否则新增文档时优先使用 createsmartcanvasby_markdown；编辑已有智能文档时使用 smartcanvas.* 系列工具
• 创建文档时支持 parentid：所有 create*bymarkdown 和 createflowchartbymermaid 工具均支持 parentid 参数，可将文档直接创建到指定目录；不填则在根目录创建
• 删除节点：deletespacenode 默认仅删除当前节点（remove_type=current），使用 all 时会递归删除所有子节点，需谨慎
• Markdown 内容使用 UTF-8 格式，特殊字符无需转义
• 幻灯片必须遵循层级结构，每页包含 2-4 个段落标题
• 分页查询每页返回 20-40 条记录，使用 has_next 判断是否有更多
• nodeid 同时也是文档的 fileid
• createflowchartby_mermaid 的 mermaid 内容必须全部使用英文
• 智能文档元素操作：Text、Heading、Task 必须挂载在 Page 下，parentid 必须为 Page 类型元素 ID；操作前先调用 smartcanvas.gettoplevelpages 获取页面结构
• 智能文档分页查询：smartcanvas.getpageinfo 使用 cursor 分页，is_over=true 表示已获取全部内容
• 智能文档删除注意：删除 Page 元素时，其下所有子元素也会被一并删除
• 智能表格操作：所有 smartsheet.* 工具都需要 fileid 和 sheetid，操作前先调用 smartsheet.listtables 获取 sheetid
• 字段类型不可更新：updatefields 时 fieldtype 不能修改，但必须传入原值
• 记录字段值格式：不同字段类型的值格式不同，详见 references/smartsheet_references.md - 字段值格式参考

问题定位指南

常见错误码及解决方案

| 错误码 | 错误类型 | 解决方案 | |--------|----------|----------| | 400006 | Token 鉴权失败 | 🔑 检查 Token 配置：确认 Header 的 key 必须使用 Authorization；同时确认 Token 值正确，可访问 https://docs.qq.com/open/auth/mcp.html 重新获取 | | 400007 | VIP权限不足 | ⭐ 立即升级VIP：访问 https://docs.qq.com/vip?immediatebuy=1?partaid=persnlspacemcp 购买VIP服务 | | -32601 | 请求接口错误 | 🔍 检查请求工具 确认调用的工具是否在工具列表中存在 | ｜ -32603 | 请求参数错误 | 🔍 检查请求参数：确认请求参数是否正确，例如file_id、content 等 |

问题排查步骤

• 检查错误信息：查看错误信息，确定错误类型，例如环境变量是否配置正确、网络问题、业务参数问题
• 检查请求参数：确认请求参数是否正确，例如file_id、content 等
• 阅读参考文档：references/ 目录下的参考文档中包含所有工具的参数说明，可帮助快速定位问题
• 获取工具列表：使用 mcporter list tencent-docs 获取所有工具列表，确认工具是否可用，检查有的参数是否正确

tencentcloud-lighthouse-skill

{baseDir}/scripts/setup.sh --check-only

{baseDir}/scripts/setup.sh --secret-id "<用户提供的SecretId>" --secret-key "<用户提供的SecretKey>"

mcporter call lighthouse. --config ~/.mcporter/mcporter.json --output json [--args '']

mcporter list lighthouse --config ~/.mcporter/mcporter.json --schema

原标题：查询所有可用地域 — 唯一不需要 Region 参数的操作
原标题：首次使用时应先调用此接口获取可用 Region 列表
mcporter call lighthouse.describe_regions --config ~/.mcporter/mcporter.json --output json

原标题：查询实例列表（Region 必填，可选参数: InstanceIds, Offset, Limit）
mcporter call lighthouse.describe_instances --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","Limit":20,"Offset":0}'
原标题：查询指定实例
mcporter call lighthouse.describe_instances --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceIds":["lhins-xxxxxxxx"]}'
原标题：启动实例
mcporter call lighthouse.start_instances --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceIds":["lhins-xxxxxxxx"]}'
原标题：获取实例登录终端地址
mcporter call lighthouse.describeinstancelogin_url --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceId":"lhins-xxxxxxxx"}'
原标题：查询所有应用镜像
mcporter call lighthouse.describeallapplications --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou"}'
原标题：BlueprintType 可选: APPOS | PUREOS | DOCKER | ALL（默认ALL）

原标题：获取监控数据（支持多指标同时查询，默认最近6小时）
mcporter call lighthouse.getmonitordata --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceId":"lhins-xxxxxxxx","Indicators":["CPU利用率","内存利用率"]}'
原标题：获取监控数据（指定时间范围）
mcporter call lighthouse.getmonitordata --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceId":"lhins-xxxxxxxx","Indicators":["公网出带宽","公网入带宽"],"StartTime":"2026-02-09 00:00:00","EndTime":"2026-02-10 00:00:00"}'
原标题：支持的监控指标（中文名称）:
原标题：CPU利用率 | 内存利用率 | 公网出带宽 | 公网入带宽
原标题：系统盘读IO | 系统盘写IO | 公网流量包
原标题：设置告警策略
原标题：Alarms 中的 Frequency/Points/Size 均为字符串类型
mcporter call lighthouse.setalertingstrategy --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceId":"lhins-xxxxxxxx","Indicator":"CPU利用率","Alarms":[{"Frequency":"300","Threshold":"80%","Level":"严重","Points":"3","Size":"60"}],"PolicyName":"CPU高负载告警"}'
原标题：Frequency(秒): "300"|"600"|"900"|"1800"|"3600"|"7200"|"10800"|"21600"|"43200"|"86400"
原标题：Level: "提示"|"严重"|"紧急"    Points: "1"-"5"    Size: "60"|"300"
原标题：服务器自检（检测网络、防火墙、存储、状态、性能）
mcporter call lighthouse.self_test --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceId":"lhins-xxxxxxxx"}'

原标题：查询防火墙规则
mcporter call lighthouse.describefirewallrules --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceId":"lhins-xxxxxxxx"}'
原标题：添加防火墙规则
mcporter call lighthouse.createfirewallrules --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceId":"lhins-xxxxxxxx","FirewallRules":[{"Protocol":"TCP","Port":"8080","CidrBlock":"0.0.0.0/0","Action":"ACCEPT","FirewallRuleDescription":"开放8080端口"}]}'
原标题：删除防火墙规则
mcporter call lighthouse.deletefirewallrules --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceId":"lhins-xxxxxxxx","FirewallRules":[{"Protocol":"TCP","Port":"8080"}]}'

原标题：在 Linux 实例上执行命令
mcporter call lighthouse.execute_command --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceId":"lhins-xxxxxxxx","Command":"uptime && df -h && free -m","SystemType":"Linux"}'
原标题：在 Windows 实例上执行命令
mcporter call lighthouse.execute_command --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InstanceId":"lhins-xxxxxxxx","Command":"Get-Process | Sort-Object CPU -Descending | Select-Object -First 10","SystemType":"Windows"}'
原标题：查询命令执行任务详情（自动轮询直到完成）
mcporter call lighthouse.describecommandtasks --config ~/.mcporter/mcporter.json --output json --args '{"Region":"ap-guangzhou","InvocationTaskId":"invt-xxxxxxxx"}'
原标题：注意: Command 最大 2048 字符，超长命令建议登录实例手动执行

curl -s "wttr.in/London?format=3"
原标题：Output: London: ⛅️ +8°C

curl -s "wttr.in/London?format=%l:+%c+%t+%h+%w"
原标题：Output: London: ⛅️ +8°C 71% ↙5km/h

curl -s "wttr.in/London?T"

curl -s "https://api.open-meteo.com/v1/forecast?latitude=51.5&longitude=-0.12¤t_weather=true"

1. CONNECT    → Authenticate with WordPress site

ANALYZE    → Load categories from site, analyze content for best match
GENERATE   → Create SEO-optimized tags based on content
CONVERT    → Transform markdown/HTML to Gutenberg blocks
PREVIEW    → Create draft and verify rendering
PUBLISH    → Publish or schedule the post
VERIFY     → Confirm live post renders correctly

from scripts.wp_publisher import WordPressPublisher
wp = WordPressPublisher(
    site_url="https://example.com",
    username="admin",
    password="xxxx xxxx xxxx xxxx xxxx xxxx"  # Application password
)
原标题：Test connection
userinfo = wp.testconnection()
print(f"Connected as: {user_info['name']}")

原标题：Get all categories from the WordPress site
categories = wp.getcategorieswith_details()
原标题：Returns list like:
原标题：[
原标题：  {'id': 1, 'name': 'Uncategorized', 'slug': 'uncategorized', 'count': 5},
原标题：  {'id': 2, 'name': 'Tutorials', 'slug': 'tutorials', 'count': 12},
原标题：  {'id': 3, 'name': 'Cloud Hosting', 'slug': 'cloud-hosting', 'count': 8},
原标题：]

原标题：Analyze content and suggest best category
suggestedcategory = wp.suggestcategory(
    content=article_content,
    title=article_title,
    available_categories=categories
)
原标题：Or let user choose from available options
print("Available categories:")
for cat in categories:
    print(f"  [{cat['id']}] {cat['name']} ({cat['count']} posts)")

原标题：Generate tags based on content analysis
tags = wp.generateseotags(
    content=article_content,
    title=article_title,
    max_tags=10
)
原标题：Returns list like:
原标题：['n8n hosting', 'workflow automation', 'self-hosted n8n', 
原标题： 'affordable hosting', 'docker deployment', 'node.js hosting']

原标题：Get or create all tags, returns list of tag IDs
tagids = wp.getorcreatetags(tags)

from scripts.contenttogutenberg import converttogutenberg
原标题：Convert markdown content
gutenbergcontent = converttogutenberg(markdowncontent)

原标题：Input markdown:
| Feature | Plan A | Plan B |
|---------|--------|--------|
| Price   | $10    | $20    |
原标题：Output Gutenberg:


  
Feature
Plan A
Plan B
  
Price
$10
$20

原标题：Create as draft first
result = wp.create_draft(
    title="Article Title",
    content=gutenberg_content,
    categories=[category_id],
    tags=tag_ids,
    excerpt="Auto-generated or custom excerpt"
)
postid = result['postid']
previewurl = result['previewurl']
editurl = result['editurl']

原标题：Fetch preview page to verify rendering
previewcontent = wp.fetchpreview(post_id)
原标题：Check for issues
issues = wp.validaterenderedcontent(preview_content)
if issues:
    print("Issues found:")
    for issue in issues:
        print(f"  - {issue}")

原标题：After preview approval, publish
result = wp.publishpost(postid)
liveurl = result['liveurl']

原标题：Full publish workflow in one call
result = wp.publish_content(
    title="Article Title",
    content=gutenberg_content,
    category_names=["Cloud Hosting"],  # By name, auto-resolves to ID
    tag_names=["n8n", "hosting", "automation"],
    status="publish",  # or "draft", "pending", "private", "future"
    excerpt="Custom excerpt for SEO",
    slug="custom-url-slug"
)

原标题：Schedule for future publication
from datetime import datetime, timedelta
publish_date = datetime.now() + timedelta(days=1)
result = wp.publish_content(
    title="Scheduled Post",
    content=content,
    status="future",
    date=publish_date.isoformat()
)

原标题：Verify the published post
verification = wp.verifypublishedpost(post_id)
print(f"Live URL: {verification['url']}")
print(f"Status: {verification['status']}")
print(f"Categories: {verification['categories']}")
print(f"Tags: {verification['tags']}")

from scripts.wp_publisher import WordPressPublisher
from scripts.contenttogutenberg import converttogutenberg
原标题：1. Connect
wp = WordPressPublisher(
    site_url="https://xcloud.host",
    username="admin",
    password="xxxx xxxx xxxx xxxx"
)
原标题：2. Load categories and select best match
categories = wp.getcategorieswith_details()
bestcategory = wp.suggestcategory(content, title, categories)
原标题：3. Generate SEO tags
tags = wp.generateseotags(content, title, max_tags=10)
原标题：4. Convert to Gutenberg
gutenbergcontent = converttogutenberg(markdowncontent)
原标题：5. Create draft and preview
draft = wp.create_draft(
    title="7 Best n8n Hosting Providers in 2026",
    content=gutenberg_content,
    categories=[best_category['id']],
    tags=wp.getorcreate_tags(tags)
)
print(f"Preview: {draft['preview_url']}")
原标题：6. After verification, publish
result = wp.publishpost(draft['postid'])
print(f"Published: {result['live_url']}")

WORDPRESSSITEURL=https://your-wordpress-site.com
WORDPRESSUSERNAME=yourusername
WORDPRESSPASSWORD=yourapplication_password

原标题：正常同步（更新已有文章）
python3 /root/.openclaw/workspace/skills/wordpress-sync/sync.py
原标题：清理所有文章后重新同步
python3 /root/.openclaw/workspace/skills/wordpress-sync/sync.py --clean