mcporter
Executes tools from any MCP server, allowing for browser control and vision analysis.
Install
mkdir -p .claude/skills/mcporter-consuelohq && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/16048" && unzip -o skill.zip -d .claude/skills/mcporter-consuelohq && rm skill.zipInstalls to .claude/skills/mcporter-consuelohq
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
--- name: mcporter description: Call MCP servers via MCPorter - browser automation, vision, and other MCP tools homepage: https://github.com/steipete/mcporter metadata: { "openclaw": { "emoji": "🧳", "requires": { "bins": ["npx"] } } }Key capabilities
- →Automate browser tasks
- →Analyze images and videos
- →Execute Playwright automation
- →List available MCP servers and tools
- →Configure new MCP servers
How it works
MCPorter acts as a bridge between OpenClaw and Model Context Protocol servers, allowing execution of various tools like browser automation or vision analysis.
Inputs & outputs
When to use mcporter
- →Automating browser tasks
- →Analyzing images via vision MCP
- →Interacting with external developer tools
About this skill
name: mcporter description: Call MCP servers via MCPorter - browser automation, vision, and other MCP tools homepage: https://github.com/steipete/mcporter metadata: { "openclaw": { "emoji": "🧳", "requires": { "bins": ["npx"] } } }
MCPorter
MCPorter is a TypeScript runtime and CLI that bridges OpenClaw to Model Context Protocol (MCP) servers. Use this skill to:
- Browser automation via chrome-devtools MCP (snapshots, clicks, forms, navigation, screenshots)
- Vision analysis via Z.AI MCP server (images, videos)
- Code automation via Playwright MCP
- Any other MCP server configured in MCPorter
Quick Start
List available MCP servers and tools
npx mcporter list
Call a specific MCP tool
# Chrome DevTools examples (✅ working - tested with GitHub.com)
npx mcporter call chrome-devtools.take_snapshot
npx mcporter call chrome-devtools.navigate_page type:url url:"https://example.com"
npx mcporter call chrome-devtools.take_screenshot format:png
npx mcporter call chrome-devtools.click uid:"element-id"
npx mcporter call chrome-devtools.fill uid:"input-id" value:"hello world"
npx mcporter call chrome-devtools.evaluate_script function:"() => document.title"
npx mcporter call chrome-devtools.list_pages
npx mcporter call chrome-devtools.list_console_messages
npx mcporter call chrome-devtools.list_network_requests
# Z.AI Vision examples (⚠️ requires Z_AI_API_KEY)
npx mcporter call zai-vision.image_analysis filePath:"/path/to/image.png"
npx mcporter call zai-vision.video_analysis filePath:"/path/to/video.mp4"
Key MCP Servers
chrome-devtools (26 tools)
Browser automation and inspection:
take_snapshot— Page snapshot (a11y tree with element UIDs)click— Click element by UIDhover— Hover over elementdrag— Drag element to drop targetfill— Fill form inputfill_form— Fill multiple form elementsnavigate_page— Navigate (url/back/forward/reload)new_page— Open new pageselect_page— Select page contextclose_page— Close pagetake_screenshot— Screenshot (PNG/JPEG/WebP)evaluate_script— Run JavaScriptpress_key— Keyboard inputresize_page— Resize viewportemulate— Network/CPU/geolocation/user agent emulationlist_console_messages— Console logslist_network_requests— Network logsget_console_message— Get specific console messageget_network_request— Get specific network requestupload_file— Upload via file inputhandle_dialog— Handle browser dialogswait_for— Wait for text/elementperformance_start_trace/performance_stop_trace— Performance tracingperformance_analyze_insight— Analyze performance insights
microsoft/playwright-mcp (22 tools)
Browser automation via Playwright (currently has connection issues).
zai-vision (requires Z_AI_API_KEY)
Image and video analysis:
ui_to_artifact— UI screenshots to code/specs/promptsextract_text_from_screenshot— OCR and text extraction from screenshotsdiagnose_error_screenshot— Analyze error screenshots and provide solutionsunderstand_technical_diagram— Diagram interpretation (architecture, flow, UML, ER)analyze_data_visualization— Charts and dashboard analysisui_diff_check— Compare two UIs to identify differencesimage_analysis— General-purpose image understanding (fallback)video_analysis— Video understanding (≤8MB, MP4/MOV/M4V)
upstash/context7 (2 tools)
Library documentation:
resolve_library_id— Get library IDget_library_docs— Fetch library docs
Configuration
View current MCP servers
npx mcporter list
Add a new MCP server
npx mcporter config add <name> <target>
# Examples:
npx mcporter config add zai-vision "npx -y @z_ai/mcp-server"
npx mcporter config add chrome-devtools "npx -y chrome-devtools-mcp@latest"
Set API keys for MCP servers that need them
export Z_AI_API_KEY=your-key-here
# Or add to shell profile (~/.zshrc, ~/.bashrc, ~/.config/fish/config.fish)
echo 'export Z_AI_API_KEY=your-key-here' >> ~/.zshrc
Notes
MCPorter auto-discovers MCP servers from:
- Cursor (~/.cursor/mcp.json)
- Claude Desktop (~/.claude.json)
- Claude Code (~/.claude-desktop/mcp.json)
- Codex (~/Library/Application Support/Code/User/mcp.json)
- VS Code
- Windsurf
- OpenCode
Workspace config
OpenClaw's MCPorter config lives at: /Users/kokayi/.openclaw/workspace/config/mcporter.json
Current configured servers:
chrome-devtools(26 tools) — ✅ Working (via stdio)zai-vision(8 tools) — ⚠️ Needs Z_AI_API_KEY
Auto-discovered servers
The following MCP servers are auto-discovered from ~/.claude.json:
chrome-devtools(from Claude Desktop) — 26 tools, browser automationupstash/context7(from Code) — 2 tools, library docs
Note: Auto-discovered servers are available via their source tools (e.g., Claude Desktop, VS Code) but may not be directly accessible through MCPorter CLI without importing them to the workspace config.
When to use MCP tools
- Browser automation — Use
chrome-devtoolsMCP tools - Vision/OCR — Use
zai-visionMCP tools (requires API key) - Complex workflows — Combine multiple MCP calls
Syntax variations (all work)
# Function-call style (recommended)
npx mcporter call 'chrome-devtools.click(uid: "element-id")'
# Flag style
npx mcporter call chrome-devtools.click uid:"element-id"
# Colon-delimited
npx mcporter call chrome-devtools.click uid:"element-id"
Troubleshooting
"tools unavailable" error
If you see "tools unavailable" for a server, it means:
- MCP server is not running (for stdio)
- Connection failed (for HTTP)
- Missing API keys or auth
Fix for microsoft/playwright-mcp
Currently shows SSE/connection errors. This is a known issue with the MCP server.
Z.AI MCP requires API key
Set environment variable:
export Z_AI_API_KEY=your-zai-api-key
Or add to MCPorter config (already configured ✅):
{
"mcpServers": {
"zai-vision": {
"command": "npx -y @z_ai/mcp-server",
"env": {
"Z_AI_API_KEY": "your-zai-api-key"
}
}
}
}
✅ Configured: Z.AI API key is set in MCPorter config. Vision tools are ready to use!
Resources
- MCPorter docs: https://github.com/steipete/mcporter
- MCP spec: https://modelcontextprotocol.io/
When not to use it
- →When the task does not involve Model Context Protocol servers
- →When the task requires direct interaction with Claude Desktop or VS Code auto-discovered servers without importing them
Prerequisites
Limitations
- →microsoft/playwright-mcp currently has connection issues
- →zai-vision requires Z_AI_API_KEY to be set
- →Auto-discovered servers may not be directly accessible through MCPorter CLI without importing them
How it compares
This skill provides a unified command-line interface to interact with diverse MCP servers, centralizing control over different automation and analysis tools.
Compared to similar skills
mcporter side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| mcporter (this skill) | 0 | 6mo | Review | Intermediate |
| browser-automation | 39 | 1mo | Review | Intermediate |
| browser-use | 64 | 3mo | Review | Intermediate |
| web-browser | 2 | 2mo | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
browser-automation
browserbase
Automate web browser interactions using natural language via CLI commands. Use when the user asks to browse websites, navigate web pages, extract data from websites, take screenshots, fill forms, click buttons, or interact with web applications. Triggers include "browse", "navigate to", "go to website", "extract data from webpage", "screenshot", "web scraping", "fill out form", "click on", "search for on the web". When taking actions be as specific as possible.
browser-use
browser-use
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, or extract information from web pages.
web-browser
mitsuhiko
Allows to interact with web pages by performing actions such as clicking buttons, filling out forms, and navigating links. It works by remote controlling Google Chrome or Chromium browsers using the Chrome DevTools Protocol (CDP). When Claude needs to browse the web, it can use this skill to do so.
autonomous-loops
Rajbharti06
Self-improvement loop patterns — eval → learn → improve cycles running without human input
gui-task
bivex
>
telegrab-fetch
saturov
Use when the user needs to download one Telegram video from a post link through this repository's telegrab CLI with deterministic preflight checks and machine-readable result fields.