Allows to interact with YouTube, including searching videos, liking, and commenting.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Pilot Browse MCP Explore twitter.com and generate a manual for posting a tweet."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP toolset + website capability learner for the browser. Agent autonomously explores websites and builds understanding manuals.
Install a skill. As the agent interacts with a website, it automatically discovers APIs and binds them with browser workflows into reusable capabilities.
Learning output is saved in website-manuals/:
website-manuals/<site>/
README.md # Root index
pages/ # Page interaction models
navigation/ # Navigation paths
workflows/
README.md # Workflow index
flows/ # Workflow JSON files
scripts/ # PAB automation scripts (.pab files)
apis/
README.md # API index (browse first)
endpoints/ # API JSON files(Using Bilibili as demo site)
Stuck on complex interactions? Record or mark elements manually to help the agent learn.
Based on the exploration manual, the agent can quickly understand websites and complete tasks faster.
Two no-LLM modes are also supported:
.pab scripts for automation tasks -- capabilities match the agent's MCP tools
Python/other scripts for specialized data collection tasks
Build various workflows, simple demo showcase.
Search a novel, save the first 5 chapters.
Generate various scripts based on the exploration manual.
For example, a novel scraper script:
I used Pilot to analyze the novel website structure, chapter API, generated scraping scripts, built Docker images, deployed 9 containers on NAS, and crawled over 4000 novels in 2 days.
Related MCP server: Cloudflare Playwright MCP
# Build
cd server && npm install && npm run build
cd extension && npm install && npm run build
# Load extension
# chrome://extensions/ -> Developer mode -> Load unpacked -> extension/dist/
# Start server
cd server && node dist/index.jsExample - fill args with the actual path to server/dist/index.js:
{
"mcpServers": {
"browser-mcp": {
"command": "node",
"args": ["/path/to/server/dist/index.js"]
}
}
}Check scripts/Skill for available skills.
Refer to agent-examples/ for ready-to-use agent workspace examples. Run your agent in one of those directories and it will automatically load the MCP config, skills, and project prompts. (Remember to update the MCP path to your actual setup.)
AI Agent (Claude Code / Pi / Codex)
|
| 1) MCP stdio protocol (JSON-RPC)
| stdin / stdout
v
MCP Server (Node.js) protocol translator
|
| 2) WebSocket :9456
v
Chrome Extension
|
| 3) Chrome API
|
v
Browser62 MCP tools -- tab management, content extraction, DOM operations, network interception, file saving, workflow recording, script automation, SQL injection detection, JS reverse
PAB scripting -- Python-like DSL for browser automation with if/for/fn support. Run from popup, no LLM needed
Element picker -- click any element on the page and tell the agent what it is
Workflow recording -- demonstrate operations to the agent, it learns and reuses
Network API toolkit -- monitor, search, inspect, replay with overrides, export code, and analyze site API structure
Token-efficient saving -- save page content directly to disk, bypassing the LLM
Shadow DOM + contenteditable
Category | Tool | What it does |
Page |
| Convert page to clean Markdown via Readability + Turndown |
| Get plain text of the page (lighter than get_html) | |
| Get raw HTML of the page (heavy, last resort) | |
| Find element by visible text, aria-label, or role | |
| Get current tab URL and title | |
| See page structure (headings, sections, buttons) | |
| Query elements by CSS selector (penetrates Shadow DOM) | |
| Execute JS in page context | |
| Extract article metadata (title, author, date, body) | |
| Extract HTML table as JSON array | |
| Extract all links from the page | |
| Extract image info (src, alt, size) | |
Actions |
| Click an element (composed:true for Shadow DOM) |
| Type text into input or contenteditable | |
| Scroll the page | |
| Wait for a given number of milliseconds | |
| Wait for an element to appear | |
Saving |
| Auto-detect main content and save to file (zero LLM tokens) |
| Extract by XPath and save to file | |
Network |
| Start intercepting requests |
| Stop monitoring (cache preserved for replay) | |
| Clear cached requests without stopping monitoring | |
| Search cached requests by keyword, method, status | |
| Get full details of a cached request (headers, body, timing) | |
| Wait for a matching request after an action (replaces fixed delay) | |
| Replay with overrides (query/headers/body) + extract JSON path | |
| Export request as curl / fetch / Python / HAR | |
| Analyze API structure of a site from cached requests | |
| Set response override rules (body, status, headers) | |
Tabs |
| List all open tabs |
| Tab management | |
Recording |
| View recordings from popup |
| Get recording details | |
| View marked elements | |
| Get marked element details | |
| List processed workflows in website-manuals | |
| Save a user-marked element to pages/ | |
| Save a processed workflow to website-manuals | |
| Generate an MCP automation script | |
| Execute an MCP automation script | |
Data |
| Read cookies (requires permission) |
| Read LocalStorage (requires permission) | |
| Take screenshot (requires permission) | |
| Permission management | |
Security |
| List security findings (with status) |
| Get details of a single finding | |
| Actively scan for SQL injection (browser-context replay) | |
| Stop scanning | |
| Advance finding status (confirm/fix/false positive) | |
| Generate a security-check.pab re-check script | |
| Custom SQL payload probe + auto verdict/extract | |
JS Reverse |
| Collect page JS files |
| AST analysis: endpoints/functions/crypto/signatures | |
| Locate a function (name/calls/crypto/callers) | |
| Associate request params with JS generator functions | |
| Query learned capability models | |
| Full reverse + save js/ + capabilities/ report |
MIT
This server cannot be installed
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
AI-powered browser automation — navigate, click, fill forms, and extract data from any website.
SaaS intelligence for AI agents. 5 unified tools cover 1,000+ services with 91-96% token savings.
Reliable web access for AI agents: smart HTTP, rotating proxies, and full-browser rendering.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/EndymionLee/PilotBrowseMCP'
If you have feedback or need assistance with the MCP directory API, please join our Discord server