Multi-session browser MCP for AI agents. 19 tools. 15 parallel sessions. Up to 10x fewer tokens.

Multi-session browser automation MCP server with stealth capabilities, supporting up to 15 parallel sessions with 10x token efficiency compared to traditional browser automation tools.

16 viewsLocal (stdio)

About leapfrog

leapfrog is a community-built MCP server published by anthonybono21-cloud that provides AI assistants with tools and capabilities via the Model Context Protocol. Multi-session browser MCP for AI agents. 19 tools. 15 parallel sessions. Up to 10x fewer tokens. It is categorized under productivity. This server exposes 19 tools that AI clients can invoke during conversations and coding sessions.

How to install

You can install leapfrog in your AI client of choice. Use the install panel on this page to get one-click setup for Cursor, Claude Desktop, VS Code, and other MCP-compatible clients. This server runs locally on your machine via the stdio transport.

License

leapfrog is released under the MIT license. This is a permissive open-source license, meaning you can freely use, modify, and distribute the software.

Tools (19)

session_create

Open a new isolated browser session with separate cookies and state

session_destroy

Close a browser session and free the slot

session_list

List all active browser sessions with URLs and idle times

session_save_profile

Save authentication state to disk for future sessions

session_list_profiles

List saved authentication profiles

Leapfrog

Leapfrog

Multi-session browser MCP for AI agents.
21 tools. 15 parallel sessions. Stealth. Humanization. Up to 10x fewer tokens.

npm i leapfrog  |  Works with Claude Code, Cursor, Windsurf


The Problem

Playwright MCP sends ~14,000 tokens for a content-heavy page like Hacker News. Most of that is noise. Your context window fills up. Your agent gets confused. You pay for it.

Leapfrog sends ~1,400 tokens. Same page. Same information. Up to 10x less noise.

┌─────────────────────────────────────────────────────┐
│  Playwright MCP                                     │
│  ████████████████████████████████████████  ~14,000   │
│                                                     │
│  Leapfrog                                           │
│  █████                                    ~1,400    │
└─────────────────────────────────────────────────────┘
          tokens per page (Hacker News, real test)

Savings range from 2-10x depending on page complexity. Content-heavy pages see the biggest wins. Dense forms see the smallest. The median across real-world sites is ~4-5x.

Quick Start

npx leapfrog --doctor   # verify everything works
npx leapfrog --config   # print MCP config to paste

Add to ~/.mcp.json (Claude Code) or your editor's MCP config:

{
  "leapfrog": {
    "command": "npx",
    "args": ["-y", "leapfrog"],
    "env": {
      "LEAP_MAX_SESSIONS": "15"
    }
  }
}

Chromium installs automatically. If it fails: npx playwright install chromium

Feature Matrix

LeapfrogPlaywright MCPagent-browser
Tokens per page~1,200-2,500~3,800-15,000~300
Parallel sessions1511
Session isolationYesNoNo
Multi-tab / popupsYesNoNo
Network interceptYesNoNo
Console captureYesYesNo
Stealth / anti-botYesNoNo
Smart wait (5 types)YesBasicNo
Crash recoveryYesNoNo
Auth profile reuseYesNoNo
SSRF protectionYesNoNo

The Ecosystem

Leapfrog uses pond metaphors to keep things memorable. Your agent is the frog.

ConceptLeapfrog termWhat it means
SessionsPondsIsolated browser contexts — cookies, storage, state
TabsLily padsWhere the frog lands within a pond
NavigateLeapJump to a URL, get a compact snapshot back
SnapshotsSurfaceWhat you see on the surface — interactive @eN refs
Network trafficRippleHTTP requests flowing under the surface
Console errorsCroakSomething went wrong in the browser
Stealth modeCamouflageAnti-bot evasion patches

All 19 Tools

Pond Management (7)

ToolWhat it does
session_createOpen a new pond — isolated cookies, state, viewport
session_destroyDrain a pond and free the slot
session_listSee all active ponds with URLs and idle times
session_save_profileSave auth state to disk for future ponds
session_list_profilesList saved auth profiles
pool_statusPool stats, memory, uptime
session_healthIs the pond healthy? Browser connected, page responsive?

Navigation & Snapshots (6)

ToolWhat it does
navigateLeap to a URL, return a compact @eN snapshot
snapshotRe-read the surface (scope with CSS selector)
actClick, fill, type, check, select, press, scroll, hover, back, forward
wait_forWait for element / text / network idle / navigation / JS expression
screenshotCapture PNG (full page or element)
extractPull text, HTML, title, URL, or evaluate JS

Tab Management (3)

ToolWhat it does
tabs_listList all pads in a pond
tab_switchHop to another pad (-1 for most recent popup)
tab_closeClose a pad (can't close the last one)

Network Intelligence (3)

ToolWhat it does
network_logSee HTTP traffic — filter by URL, method, status, content-type
console_logRead browser console output, filtered by level
network_interceptBlock, mock, or log requests by URL pattern

Environment Variables

VariableDefaultDescription
LEAP_MAX_SESSIONS15Max concurrent sessions
LEAP_IDLE_TIMEOUT300000Session idle timeout in ms (5 min)
LEAP_HEADLESStrueSet false to watch the browser
LEAP_ALLOW_JStrueAllow JS evaluation in extract and wait_for
LEAP_STEALTHtrueStealth mode (anti-bot evasion)
LEAP_LOG_LEVELinfodebug / info / warn / error

Tests

 74 passing across 5 suites

Session management, snapshot engine, network intelligence, tab management, security (SSRF, URL scheme blocking, path traversal).

npm test

Requirements

  • Node.js >= 18
  • Chromium (auto-installed via Playwright)

License

MIT

Alternatives

Related Skills

Browse all skills
ai-assisted-development

Leveraging AI coding assistants and tools to boost development productivity, while maintaining oversight to ensure quality results.

4
teams-channel-post-writer

Creates educational Teams channel posts for internal knowledge sharing about Claude Code features, tools, and best practices. Applies when writing posts, announcements, or documentation to teach colleagues effective Claude Code usage, announce new features, share productivity tips, or document lessons learned. Provides templates, writing guidelines, and structured approaches emphasizing concrete examples, underlying principles, and connections to best practices like context engineering. Activates for content involving Teams posts, channel announcements, feature documentation, or tip sharing.

4
cto-engineering-metrics

Expert methodology for defining, tracking, and interpreting engineering performance metrics including DORA, team health, productivity, and executive reporting.

4
personal-assistant

This skill should be used whenever users request personal assistance tasks such as schedule management, task tracking, reminder setting, habit monitoring, productivity advice, time management, or any query requiring personalized responses based on user preferences and context. On first use, collects comprehensive user information including schedule, working habits, preferences, goals, and routines. Maintains an intelligent database that automatically organizes and prioritizes information, keeping relevant data and discarding outdated context.

3
productivity-helper

Boost your productivity with automated task management

2
cursor-local-dev-loop

Optimize local development workflow with Cursor. Triggers on "cursor workflow", "cursor development loop", "cursor productivity", "cursor daily workflow". Use when working with cursor local dev loop functionality. Trigger with phrases like "cursor local dev loop", "cursor loop", "cursor".

2