Which integrations are available for this server?

Allows GitHub Copilot to control a browser for web automation tasks such as navigation, clicking, form filling, and screenshots.

How do I use IRA-RESEARCHER?

This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.

IRA-RESEARCHER

by neuralnexustech

Overview Schema Source Related Servers Discussions Admin

JavaScript

Local

🔬 IRA-RESEARCHER — Browser Automation MCP Server

MCP browser server for Claude, Cursor, Copilot, OpenCode, and any MCP-compatible AI tool. 40 tools, ghost panel, stealth mode, full DevTools access. No API keys needed.

Powered by Neural Nexus Tech

Tags: mcp server mcp browser browser automation puppeteer mcp claude browser cursor mcp ai agent browser mcp tools headless browser web scraping mcp devtools mcp ghost panel

✨ What is this?

IRA-RESEARCHER is an MCP (Model Context Protocol) server that gives AI assistants full control of a Chrome browser. It works with Claude Desktop, Claude Code, Cursor, VS Code Copilot, OpenCode, Windsurf, Aider, Cline, and any tool that supports MCP.

40 tools — navigate, click, type, screenshot, read pages, inspect elements, run JavaScript, intercept requests, and more
👻 Ghost panel — real-time overlay showing every action in the browser
🛡️ Stealth mode — anti-detection evasions to avoid bot blocks
🔧 Full DevTools — console, network, performance, storage, accessibility audit
🏃 Auto browser launch — opens Chrome automatically, no extension needed
🔑 Zero API keys — runs locally, no cloud dependency

Related MCP server: Ghost Browser MCP

📋 Quick Start (2 steps)

Option A: Run from npm (easiest)

npm install -g ira-researcher

Then add to your MCP config:

{
  "mcpServers": {
    "ira-researcher": {
      "command": "ira-researcher",
      "env": {
        "IRA_HEADLESS": "false",
        "IRA_WINDOW": "maximize",
        "IRA_GHOST": "true"
      }
    }
  }
}

Option B: Run from source

git clone https://github.com/neuralnexustech/IRA-RESEARCHER.git
cd IRA-RESEARCHER
npm install
npx puppeteer browsers install chrome

Then add to your MCP config (replace path):

{
  "mcpServers": {
    "ira-researcher": {
      "command": "node",
      "args": ["C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"],
      "env": {
        "IRA_HEADLESS": "false",
        "IRA_WINDOW": "maximize",
        "IRA_GHOST": "true"
      }
    }
  }
}

Add to Any MCP Client

.cursor/mcp.json — add to the mcpServers object:

{
  "mcpServers": {
    "ira-researcher": {
      "command": "node",
      "args": ["C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"],
      "env": {
        "IRA_HEADLESS": "false",
        "IRA_WINDOW": "maximize",
        "IRA_GHOST": "true"
      }
    }
  }
}

Run in terminal:

claude mcp add ira-researcher node C:\path\to\IRA-RESEARCHER\src\index.js --env IRA_HEADLESS=false --env IRA_WINDOW=maximize --env IRA_GHOST=true

Or manually edit ~/.claude/mcp.json:

{
  "mcpServers": {
    "ira-researcher": {
      "command": "node",
      "args": ["C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"],
      "env": {
        "IRA_HEADLESS": "false",
        "IRA_WINDOW": "maximize",
        "IRA_GHOST": "true"
      }
    }
  }
}

~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "ira-researcher": {
      "command": "node",
      "args": ["C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"],
      "env": {
        "IRA_HEADLESS": "false",
        "IRA_WINDOW": "maximize",
        "IRA_GHOST": "true"
      }
    }
  }
}

Project root opencode.json:

{
  "mcp": {
    "ira-researcher": {
      "command": "node",
      "args": ["C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"],
      "env": {
        "IRA_HEADLESS": "false",
        "IRA_WINDOW": "maximize",
        "IRA_GHOST": "true"
      }
    }
  }
}

.vscode/mcp.json:

{
  "servers": {
    "ira-researcher": {
      "command": "node",
      "args": ["C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"],
      "env": {
        "IRA_HEADLESS": "false",
        "IRA_WINDOW": "maximize",
        "IRA_GHOST": "true"
      }
    }
  }
}

VS Code Settings → Extensions → Cline → MCP Servers → Add:

{
  "ira-researcher": {
    "command": "node",
    "args": ["C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"],
    "env": {
      "IRA_HEADLESS": "false",
      "IRA_WINDOW": "maximize",
      "IRA_GHOST": "true"
    }
  }
}

.windsurfrules or MCP settings:

{
  "mcpServers": {
    "ira-researcher": {
      "command": "node",
      "args": ["C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"],
      "env": {
        "IRA_HEADLESS": "false",
        "IRA_WINDOW": "maximize",
        "IRA_GHOST": "true"
      }
    }
  }
}

.aider.conf.yml:

mcp-servers:
  ira-researcher:
    command: node
    args:
      - "C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"
    env:
      IRA_HEADLESS: "false"
      IRA_WINDOW: "maximize"
      IRA_GHOST: "true"

Or via CLI:

aider --mcp-server ira-researcher node C:\path\to\IRA-RESEARCHER\src\index.js

~/.config/zed/settings.json → add to context_servers:

{
  "context_servers": {
    "ira-researcher": {
      "command": "node",
      "args": ["C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"],
      "env": {
        "IRA_HEADLESS": "false",
        "IRA_WINDOW": "maximize",
        "IRA_GHOST": "true"
      }
    }
  }
}

Any MCP-compatible tool uses this JSON:

{
  "mcpServers": {
    "ira-researcher": {
      "command": "node",
      "args": ["C:\\path\\to\\IRA-RESEARCHER\\src\\index.js"],
      "env": {
        "IRA_HEADLESS": "false",
        "IRA_WINDOW": "maximize",
        "IRA_GHOST": "true"
      }
    }
  }
}

3. Use

The 43 tools appear automatically: ira_navigate, ira_click, ira_screenshot, etc.

🧰 Tools (43)

Category	Tools
Navigation	`ira_navigate`, `ira_go_back`, `ira_go_forward`, `ira_reload`
Interaction	`ira_click`, `ira_type`, `ira_hover`, `ira_drag`, `ira_upload`, `ira_select`, `ira_keyboard`
Vision & Reading	`ira_screenshot`, `ira_element_screenshot`, `ira_extract_images`, `ira_get_state`, `ira_read_page`, `ira_find`, `ira_extract_text`, `ira_extract_table`, `ira_get_html`
Tabs	`ira_tabs`, `ira_switch_tab`, `ira_close_tab`, `ira_new_tab`
Debug & DevTools	`ira_console`, `ira_network`, `ira_javascript`, `ira_inspect_element`, `ira_get_styles`, `ira_performance`, `ira_storage`, `ira_network_timing`, `ira_audit_accessibility`
Utility	`ira_wait`, `ira_scroll`, `ira_set_viewport`, `ira_cookies`, `ira_pdf`, `ira_intercept`
Instance Management	`ira_shutdown`, `ira_instances`, `ira_kill_instance`, `ira_kill_all`
Health & Status	`ira_health`

⚙️ Environment Variables

Variable	Default	Description
`IRA_HEADLESS`	`false`	Run browser in headless mode
`IRA_WINDOW`	`auto`	Window sizing: `maximize`, `auto` (detect screen), or `WxH` like `1280x800`
`IRA_GHOST`	`true`	Enable ghost overlay UI
`IRA_STEALTH`	`true`	Enable anti-detection measures
`IRA_PROXY`	``	Proxy URL or comma-separated pool
`IRA_PROXY_ROTATE`	`false`	Auto-rotate proxies randomly
`IRA_DEBUG`	`false`	Enable verbose debug logging

Window visibility: Set IRA_HEADLESS=false + IRA_WINDOW=maximize to open a maximized, focused Chrome window on Windows. The browser auto-detects your screen work area and uses --start-maximized to bypass OS focus-steal blocks.

👻 Ghost Effects

When enabled, a floating panel appears on the right side of the browser showing all actions in real-time:

▶️ Action (green) — navigation, clicks
💭 Thought (orange) — reasoning
✅ Success (green) — completed actions
❌ Error (red) — failures
🟠 Click ripples — orange expanding circles at click points
🟢 Typing glow — green glow on active input fields
🔴 Drag arrows — animated SVG path from start to end
📸 Screenshot flash — brief white flash

Toggle panel: Ctrl+Shift+I

📁 Project Structure

IRA-RESEARCHER/
├── src/
│   ├── index.js              ← Entry point (43 tools, auto-recovery)
│   ├── pipeline.js           ← Middleware: logging, timing, error wrapping
│   ├── browser.js            ← Single Chrome instance + reuse + registry
│   ├── utils.js              ← Shared helpers (textResult, ghost effects)
│   ├── tools/                ← 43 tools (10 files)
│   │   ├── navigation.js     ← 4 tools
│   │   ├── interaction.js    ← 7 tools
│   │   ├── vision.js         ← 3 tools
│   │   ├── reading.js        ← 6 tools
│   │   ├── tabs.js           ← 4 tools
│   │   ├── debug.js          ← 3 tools
│   │   ├── devtools.js       ← 6 tools
│   │   ├── utility.js        ← 7 tools (wait, scroll, set_viewport, cookies, pdf, intercept, shutdown)
│   │   ├── selectors.js      ← Shared element selectors
│   │   └── status.js         ← 1 tool (ira_health)
│   └── ghost/index.js        ← Ghost effects
├── package.json
└── README.md

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityNo data

ResponsivenessOutdated data

Resources

Related MCP Connectors

Irmu — Web Data MCP
io.github.polgoide
Web scraping, browser automation, screenshots and AI data extraction for AI agents via MCP.
Browser Automation Web Scraping Agent Orchestration
1
https://app.irmu.com/mcp
browser-useA
com.browser-use
Browser agents and stealth cloud browsers for AI: run web tasks, fill forms, extract data.
Agent Orchestration Browser Automation Web Scraping
7170,259
https://api.browser-use.com/v3/mcp
Skyvern
io.github.Skyvern-AI
AI-powered browser automation — navigate, click, fill forms, and extract data from any website.
Browser Automation Web Scraping App Automation
223,398
https://api.skyvern.com/mcp/
Browserless
io.browserless
Official Browserless MCP server: browser automation, scraping, screenshots and PDFs for AI agents.
Browser Automation Web Scraping Agent Orchestration
119
https://mcp.browserless.io/mcp

Related MCP Servers

MCP Camoufox
Browser Automation Web Scraping
RobithYusuf
A
license
B
quality
C
maintenance
A stealth browser MCP server with 69 tools for undetectable web automation. It bypasses bot detection systems like Cloudflare while providing full browser control for tasks like form filling, data scraping, and session management.
Updated 9 days ago (2026-11-04 06:51 UTC)
133
368 npm
23
MIT
Ghost Browser MCP
BU3NO77XL
-
license
Not graded
quality
Not graded
maintenance
Undetectable browser automation server for MCP-compatible AI agents, offering 225 tools across 32 sections to navigate, extract, clone pages, and bypass antibot systems like Cloudflare.
Updated 6 months ago (2026-05-07 04:29 UTC)
1
-
Browser Automation MCP
Browser Automation Autonomous Agents
PopoverAI
A
license
A
quality
B
maintenance
MCP server for AI browser automation with tools for navigation, actions, data extraction, and scripting, supporting local and cloud execution.
Updated a month ago (2026-09-30 19:20 UTC)
14
496 npm
3
Apache 2.0
wraith-mcp
Browser Automation AI & Machine Learning
Koreahwan
A
license
A
quality
C
maintenance
An AI-native stealth browser MCP server that uses AI vision navigation and bot detection bypass, enabling natural language browser automation.
Updated 5 months ago (2026-05-30 16:49 UTC)
17
MIT