Skip to main content
Glama
Sign In
deenesjakoruzh
browser-firefox-mcp

browser-firefox-mcp

by ironessi
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

browser-firefox-mcp

Security-focused Firefox browser MCP server for penetration testing agents. Built on Playwright with a Python-native API designed for vulnerability scanning, session manipulation, network inspection, and reconnaissance.

Features

  • Firefox-first: Uses Playwright's native Firefox engine for accurate rendering (especially useful for anti-bot detection)

  • Security tooling: XSS payload injection, form enumeration, DOM analysis, sensitive data leak detection

  • Session management: Cookie export/inject, token extraction, storage manipulation

  • Network inspection: Request/response interception and capture via CDP

  • Reconnaissance: Technology stack fingerprinting, framework detection, metadata extraction

  • MCP stdio transport: Works with cyberstrikeai, Claude Code, Cursor, and any MCP-compatible client

Related MCP server: mcp-browser-automation

Quick Start (Kali Linux)

git clone https://your-repo/browser-firefox-mcp.git
cd browser-firefox-mcp
uv venv && source .venv/bin/activate
uv pip install -e .
playwright install firefox
python server.py  # or: browser-firefox-mcp

Integration with cyberstrikeai

Add to your ~/.config/cyberstrikeai/config.json (or equivalent):

{
  "mcp_servers": {
    "firefox-browser": {
      "command": "python",
      "args": ["server.py"],
      "cwd": "/path/to/browser-firefox-mcp"
    }
  }
}

The agent will automatically discover tools: navigate, click, fill, screenshot, inject_payload, scan_forms, capture_network, get_session_info, recon_technology_stack, etc.

Architecture

┌──────────────────┐     JSON-RPC (stdio)     ┌──────────────────┐
│  cyberstrikeai   │ ◄──────────────────────► │     server.py    │
│   (agent)        │                          │ (MCP stdio)      │
└──────────────────┘                          └────────┬─────────┘
                                                       │
                                          ┌────────────▼─────────┐
                                          │  core/ modules        │
                                          │  engine | page        │
                                          │  security | session   │
                                          │  network | recon      │
                                          └────────────┬─────────┘
                                                       │
                                               ┌───────▼───────┐
                                               │ Firefox (headless) │
                                               └─────────────────┘

License

MIT

Available Tools

40 tools
block_urlsC

Block navigations matching glob-style URL patterns.

Inputs: patterns (required).

clear_cookiesC

Clear all cookies for the current context.

Inputs: no named parameters.

clickC

Click an element identified by CSS selector.

Inputs: selector (required), timeout.

click_textA

Click the first node whose visible text contains the given text.

Inputs: text (required), timeout.

close_browserA

Close the browser and release resources.

Inputs: no named parameters.

dom_analysisB

Extract structural summary of the DOM tree.

Inputs: no named parameters.

enumerate_formsA

List all forms with actions, methods, and field details.

Inputs: no named parameters.

enumerate_linksC

List all hyperlinks on the page.

Inputs: relative_only.

export_stateA

Save full browser state (cookies + storage) to a file.

Inputs: path (required).

fill_inputB

Fill a form field. Fires framework-compatible input/change events.

Inputs: value (required), selector (required), timeout, clear_first.

find_reflectionA

Check whether text was reflected in the current page DOM.

Inputs: text (required).

find_sensitive_dataA

Scan page HTML and text for leaked secrets, keys, tokens.

Inputs: no named parameters.

get_cookiesA

List all cookies, optionally filtered by domain patterns.

Inputs: domains.

get_network_summaryC

Get a filtered summary of captured network traffic.

Inputs: status_filter, resource_types.

get_session_infoA

Full session snapshot: cookies + storage + detected tokens.

Inputs: no named parameters.

get_storage_dataA

Dump localStorage and sessionStorage contents.

Inputs: no named parameters.

get_textC

Extract visible text under the given CSS selector.

Inputs: selector.

get_url_paramsA

Extract and list all URL query parameters.

Inputs: no named parameters.

go_backC

Navigate back in history.

Inputs: no named parameters.

go_forwardC

Navigate forward in history.

Inputs: no named parameters.

import_stateB

Restore browser state from a previously saved file.

Inputs: path (required).

inject_payloadC

Inject a single custom payload into a target element and verify reflection.

Inputs: payload (required), selector (required).

inject_xss_payloadC

Inject XSS probes into a target input/textarea and check for reflection.

Inputs: selector (required), payloads, clear_first.

js_evalC

Execute JavaScript expression and return result.

Inputs: expression (required).

list_contextsA

List all active browser contexts.

Inputs: no named parameters.

navigateA

Navigate to a URL. Returns final url, title, and HTTP status.

Inputs: url (required), timeout_ms.

new_contextA

Create a new isolated browser context (separate cookie jar).

Inputs: proxy.

press_keyA

Press a keyboard key or combination (e.g. 'Enter', 'Control+a').

Inputs: key (required).

read_pageA

Read the full page: DOM HTML length + title + metadata summary.

Inputs: no named parameters.

recon_technology_stackB

Detect technologies/frameworks/CMS used on the current page.

Inputs: no named parameters.

reload_pageD

Reload the current page.

Inputs: no named parameters.

screenshotA

Take a screenshot. Returns base64 string, or saves to path.

Inputs: path, full_page.

security_headers_checkB

Check for common security headers and protections.

Inputs: no named parameters.

set_cookieB

Set a single cookie. Useful for session token replay.

Inputs: name (required), value (required), path, domain, +3 more.

set_cookiesA

Bulk-inject cookies (e.g., from Burp export).

Inputs: cookies (required).

set_storage_itemC

Write a key-value pair to localStorage or sessionStorage.

Inputs: key (required), value (required), store.

start_network_captureA

Begin capturing all HTTP(S) requests/responses on the current page.

Inputs: no named parameters.

stop_network_captureB

Stop capturing and report totals.

Inputs: no named parameters.

upload_fileB

Upload file(s) to a file input element.

Inputs: selector (required), file_paths (required).

wait_for_selectorC

Wait for an element to appear. Returns True when found.

Inputs: selector (required), timeout_ms.

Tool Schema Changelog

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

Loading tool schema changelog…

    Deploy Server
    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

    TDQS

    C2.6/5.0

    Scored across 40 tools

    Disambiguation3/5

    Several tools overlap in purpose, such as inject_xss_payload vs inject_payload and find_reflection vs find_sensitive_data, though descriptions help distinguish them. Most tools are distinct, but the boundaries are not always crystal clear.

    Naming Consistency3/5

    Naming is mostly verb_noun with underscores, but there are deviations like 'dom_analysis' and 'security_headers_check' that reverse the pattern, and 'new_context' is not a clear verb. Mixed conventions remain readable but lack strict consistency.

    Tool Count2/5

    With 40 tools, the server exceeds the recommended range for a focused MCP server. Many tools are highly specialized (e.g., network capture, security scanning) and contribute to a heavy surface, making it feel over-scoped.

    Completeness4/5

    The toolset covers a wide range of browser automation and security testing operations, including navigation, interaction, cookies, storage, network capture, and security checks. Minor gaps like hover/focus or tab management exist, but overall it's comprehensive for its domain.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Resources

    Related MCP Connectors

    Related MCP Servers

    • browser-mcp-server

      ricardodeazambuja
      A
      license
      B
      quality
      D
      maintenance
      A universal browser automation MCP server using Playwright, enabling programmatic control of Chrome with 63 tools for navigation, interaction, media control, and CDP-based diagnostics.
      63
      28 npm
      3
      MIT

    • mcp-browser-automation

      ritesh-jain
      A
      license
      B
      quality
      D
      maintenance
      An MCP server for generic browser automation using Playwright. Enables MCP clients to navigate pages, inspect elements, execute JavaScript, capture screenshots, and monitor console logs and network traffic via a headless Chromium instance.
      7
      27 npm
      ISC