Skip to main content
Glama
xav-ie

Browser Session MCP

Related Servers

Alternatives to Browser Session MCP

  • A
    license
    A
    quality
    A
    maintenance
    Umbra is the trust layer for AI-generated code. It scores any repo 0-100 with file:line evidence (security, slop, Docker-verified build/boot, and claim receipts that catch agents lying about tests), and guards agent file writes mid-stream blocking leaked keys, alg:none JWTs, and git-hook plants before they land. Tools: scan_repo, guard_content, get_score.
    3
    31 npm
    MIT
  • A
    license
    Not graded
    quality
    A
    maintenance
    A local MCP server that lets AI agents and human operators drive isolated Chromium profiles with their own cookies, fingerprints, and proxies, exposing tools for navigation, clicking, typing, extraction, screenshots, JavaScript evaluation, tabs, cookies, and waits. Because session and login state persist per profile, agents can carry out real authenticated workflows across sessions.
    MIT
  • A
    license
    Not graded
    quality
    A
    maintenance
    Enables MCP-compatible AI assistants to drive the user's real, already-logged-in Chrome browser locally, so they can read and summarize pages, click, type, fill and submit forms, navigate tabs, and capture screenshots with full access to existing sessions. It also exposes cookies, storage, console, network capture, and emulation capabilities without opening a remote-debugging port or sending data off the machine.
    1 npm
    ISC

Related Servers

  • F
    license
    Not graded
    quality
    A
    maintenance
    Exposes a local MCP endpoint that lets AI agents and human operators drive isolated Chromium profiles — each with its own cookies, fingerprint, and proxy — through tools for navigation, clicking, typing, extraction, screenshots, JS evaluation, tabs, and cookie management. Persistent per-profile session state keeps logins and context alive across launches and across AI sessions, so agents can complete real authenticated workflows.
    -
  • A
    license
    Not graded
    quality
    D
    maintenance
    Stealth Chrome MCP server for AI agents providing 77 tools, perfect bot detection scores, proxy pools, encrypted sessions, and live human handoff via Cloudflare tunnel.
    MIT
  • A
    license
    Not graded
    quality
    C
    maintenance
    Gives AI agents a single shared, logged-in Chrome they can drive over MCP or plain HTTP: open and read tabs, click, type, upload files, run short Playwright scripts, hand off whole web tasks to a cheaper model, and keep per-site memory. Agents work in background windows they own without stealing the user's mouse or closing their tabs.
    MIT
  • A
    license
    Not graded
    quality
    A
    maintenance
    A local MCP server that lets AI assistants drive your real, already-logged-in Chrome through a browser extension, so agents can read, click, type, navigate, screenshot, inspect network traffic, and manage cookies or storage on pages that require existing sessions. It works without opening a remote-debugging port and keeps all communication on your machine.
    1 npm
    ISC
  • A
    license
    Not graded
    quality
    A
    maintenance
    A local MCP server that lets AI coding tools drive your own already-logged-in Chrome through a browser extension, rather than a headless or cookie-copied instance. It exposes 25 tools for snapshotting pages, clicking, typing, reading content, taking screenshots, recording network traffic, and reading cookies or storage, so agents can act on authenticated sites without triggering bot detection.
    1 npm
    ISC
  • A
    license
    Not graded
    quality
    A
    maintenance
    An MCP server that gives AI agents faithful control of real, logged-in Chrome profiles via a secure local bridge and extension, with exact multi-profile routing and no debug profiles or remote debugging ports.
    5
    MIT

TDQS

A4.5/5.0

Scored across 30 tools

Disambiguation5/5

Every tool has a sharply distinct purpose, and the descriptions explicitly cross-reference siblings to disambiguate (navigate vs new_page, list_pages vs list_browser_sessions vs list_visits vs list_browser_states, take_screenshot vs take_snapshot vs evaluate, click vs evaluate-based clicks). Even the closely-related logging and state tools each carve out a clear, non-overlapping role. An agent could reliably select the right tool without guessing.

Naming Consistency4/5

Names are uniformly snake_case and mostly use a predictable verb_noun pattern (open_browser_session, close_tab, list_pages, take_screenshot, save_browser_state, request_human_takeover). A few break the pattern as bare verbs (navigate, click, type, scroll, evaluate) and 'new_page' should arguably be 'open_page' for symmetry with open_browser_session/close_tab. Still highly readable and internally coherent.

Tool Count3/5

30 tools is heavy and stretches past a comfortable surface, though the breadth reflects genuine scope (sessions, tabs, interaction, perception, logging, profiling, stealth, takeover, cookie state). Each tool earns its place individually, but the set is large enough that an agent must navigate many options for a single task. Borderline over-scoped rather than well-scoped.

Completeness4/5

Covers the core lifecycle well: session/tab management, navigation, trusted input, multiple perception modes, JS eval, console/network logs, profiling, stealth, human takeover, and cookie persistence. Gaps remain around browser history (back/forward), file upload/download, drag, iframe targeting, and non-cookie storage (localStorage/auth beyond cookie snapshots), but these are minor workarounds rather than dead ends.

Maintenance

ActivityMaintained
ResponsivenessNo issues