Show HN: VibeBrowser – Give your AI agent your real logged-in browser via MCP
VibeBrowser introduces an MCP (Model Context Protocol) tool that connects AI agents directly to a user's real, logged-in Chrome browser session without requiring debug modes or separate browser instances. It enables multiple AI agents to control the same browser simultaneously, supporting automation across authenticated websites, extensions, and internal tools. The system works locally or via an internet-exposed relay, offering 25+ tools for navigation, interaction, Google Workspace, and credential management. It is free to use, integrates with popular AI agents, and emphasizes security with a secret vault that avoids exposing credentials to LLMs.
- ▪VibeBrowser MCP connects AI agents to a user's actual logged-in Chrome browser, preserving cookies, tabs, and extensions.
- ▪It supports multi-agent control through a local or internet-accessible relay, allowing multiple AI agents to share one browser session.
- ▪The toolset includes 25+ capabilities across navigation, interaction, Google Workspace, and credential management, with markdown-based page snapshots for lower token usage.
- ▪Unlike Playwright or Chrome DevTools MCP, VibeBrowser does not require launching a separate browser or enabling debug ports.
- ▪The MCP package is open source, though the browser extension itself is not, and it integrates with agents like Claude, Cursor, and VS Code.
Full article excerpt tap to expand
Real Browser MCPGive Any AI Agent|a real browserYour real logged-in browser — connected directly to your AI agent. No Browserbase account. No stealth mode. No per-hour billing. Just your existing Chrome session, exposed as MCP.Install in ChromeInstall in Claude CodeCLIclaude mcp add --transport stdio --scope user vibe -- npx -y --package @vibebrowser/mcp@latest vibebrowser-mcpOr add to project-level .mcp.jsonRequires Node.js and the Vibe Browser extensionPublished MCP binaries: vibebrowser-mcp and vibe-mcp. For OpenClaw-style CLI flows, see the separate OpenClaw page.25+ browser toolsMulti-agent relayInternet-exposed relayNo debug permissionsOpen source MCP packageFree — no Browserbase account neededWorks with any MCP-compatible agentCClaude CodeOOpenCodeCCursorCClaude DesktopVVS CodeWWindsurfGGemini CLICCodexCapability surfacesWhat agents can actually do through Vibe Browser MCPPackage the browser, workspace, security, and coordination layers as explicit surfaces. That is the useful lesson from infrastructure products like Tavily, translated to a real-browser MCP.Browser controlSnapshotsWorkspaceSecrets + memoryCoordinationBrowser controlDrive the live browser session your operator already trustsOpen tabs, navigate, click, fill, drag, type, and scroll inside the same logged-in browser profile instead of a disposable automation runtime.Why teams reach for this surfaceThis is the right surface when your agent needs real UI state, installed extensions, existing cookies, and human-visible execution.Works on authenticated SaaS, portals, and internal toolsKeeps the human in the same browser context as the agentAvoids extra browser launch flags or CDP ceremonyRepresentative toolsnavigate_pageclickfill_formpress_keydragPackaging that technical buyers understandTavily makes its product legible by naming a small set of surfaces clearly. The equivalent move for Vibe is not “search” or “crawl”; it is browser control, snapshots, workspace actions, secrets, and agent coordination.Proof points already present in this route25+published tools across browser, workspace, security, and agent orchestrationLocal + remoterelay paths for laptop agents and internet-reachable runnersMulti-agentone real browser session can be shared by multiple cooperating agentsReal sessionscookies, tabs, extensions, and authenticated apps stay intactHow Vibe Browser MCP comparesVibe Browser MCP is the only browser MCP that uses your real browser with all your logged-in sessions — no debug ports, no separate browser instance.FeatureVibe Browser MCPPlaywright MCPDevTools MCPBrowser MCPUses your logged-in browserPlaywright supports this via its extension mode, but defaults to launching a separate browser instanceNo debug port requiredPlaywright's default mode requires --remote-debugging-port; extension mode avoids this but adds token auth setupMulti-agent supportMultiple AI agents control the same browser simultaneously via relay daemonInternet-exposed relayExpose your relay to the internet so remote agents can connect to your local browser from anywhereGoogle Workspace integrationGmail search/send/draft/thread/message plus Calendar view/create/delete — 8 native toolsCredential vaultSecure password manager that never exposes secrets to the LLMSub-agent orchestrationSpawn sub-agents with isolated context and parallel tool executionStandalone AI browserAlso works as a standalone AI co-pilot directly in your browserOpen sourceVibe's npm package (@vibebrowser/mcp) ships…
This excerpt is published under fair use for community discussion. Read the full article at Vibe Browser.
Discussion
0 commentsMore from Vibe Browser
