sa2web-mcp (sa2web/sa2web-mcp) is an MCP server listed on the M8ven Trust Index. It scores 50 out of 100, grade D. It declares 40 tools. No publisher has claimed this listing.

Caution

50/100

sa2web-mcp

Item: sa2web-mcp
Rating: 50
Author: M8ven

Exposes a remote browser as MCP tools via Playwright, enabling AI agents to navigate and interact with web pages through DOM snapshots, clicks, typing, and form operations.

→ sa2web/sa2web-mcp · listed on glama

Caution. Specific findings reduced this grade. They are listed on the page. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.

How we verified

Code Verified⚡ Live Monitored: not connected

Verified is a snapshot. Live keeps it current, and builds your track record.

⚡ Connect GitHub → continuous verification on every push why connect →

Who stands behind it

sa2web

Source: Glama

Is this your MCP?

Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →

⚡ Connect GitHub for Live Monitored

Install from

Glamanot scored here

The grade above is for the source repository. Registries can serve a different version, so we mark the ones we were not able to read.

// key findings

⚠️

Known vulnerabilities in dependencies: 4 high

Affects packages this MCP installs at runtime. Upgrade or remove the affected dependency.

✅

No credential exfiltration, no sensitive file access, no obfuscation

Static analysis found nothing flowing your secrets to unexpected places.

✅

Open source with a license and README

Anyone can audit the code, the license is declared, and the publisher documents what it does.

// tools this server exposes40 tools

These names and descriptions are the publisher's own, read from the source code. We print them as written. Our assessment is the findings above, not this list.

sa2_help

Start here. Explains the simplest SA2 remote-browser workflow and which tools to call next.

sa2_list_available_targets

Use this first when the user wants to open a SaaS site, workspace, inner site, or asks what saved targets are available. Returns stable targetId values for sa2_open_target.

sa2_open_target

Open any SA2 target with one tool. For public websites use type=cloud and url. Saved targets can be opened by targetId or by type plus name; name matching is case-insensitive.

browser_open_login

Advanced compatibility tool. Open the configured remote browser login URL from SA2_LOGIN_URL; most agents should prefer sa2_open_target, which logs in automatically.

browser_navigate

Advanced compatibility tool. Navigate to a public URL through cloud browsing. Prefer sa2_open_target with type=cloud for simpler agent behavior.

browser_navigate_back

Go back by clicking the remote browser toolbar button #btn-back.

browser_navigate_forward

Go forward by clicking the remote browser toolbar button #btn-forward.

browser_reload

Reload by clicking the remote browser toolbar button #btn-reload.

browser_list_saas_sites

Advanced compatibility tool. List all SaaS sites from /api/v1/home/saas. Prefer sa2_list_available_targets.

browser_list_proxies

Advanced compatibility tool. List cloud browsing proxy options from /api/v1/home/freebrowse.

browser_enter_saas_site

Advanced compatibility tool. Enter a SaaS site by saasList id or name. Prefer sa2_open_target with a targetId.

browser_list_workspaces

Advanced compatibility tool. List all workspace sites and accounts from /api/v1/home/multis. Prefer sa2_list_available_targets.

browser_enter_workspace

Advanced compatibility tool. Enter a workspace account by accountId or by siteId + username. Prefer sa2_open_target with a targetId.

browser_list_inner_sites

Advanced compatibility tool. List all inner sites from /api/v1/home/innerSites. Prefer sa2_list_available_targets.

browser_enter_inner_site

Advanced compatibility tool. Enter an inner site by id or name. Prefer sa2_open_target with a targetId.

browser_snapshot

browser_extract_text

browser_click

Click an element by snapshot ref, role/name, text, selector, or x/y coordinates.

browser_type

Type or fill text into an element by snapshot ref, role/name, selector, or the focused element. Textareas and contenteditable rich-text editors preserve multiline input; both actual newlines and escaped \n sequences are accepted.

browser_press

Press a keyboard key, such as Enter, Escape, Tab, ArrowDown, or Control+A.

browser_press_key

Press a keyboard key, such as Enter, Escape, Tab, ArrowDown, or Control+A.

browser_wait

Wait for time, text, selector, or URL substring.

browser_scroll

Scroll the target page or a scrollable element. With no target or movement arguments, scrolls the root page down 600px. Use ref/selector/role/name/text with mode=intoView to reveal an element, mode=by for relative movement, or mode=to/top/bottom for an absolute position.

browser_hover

Hover over an element by snapshot ref, role/name, text, or selector.

browser_drag

Drag and drop from one element to another by ref or selector.

browser_fill_form

Fill multiple form fields at once.

browser_select_option

Select dropdown option values by ref or selector.

browser_check

Check a checkbox or radio by ref, selector, role/name, or text.

browser_uncheck

Uncheck a checkbox by ref, selector, role/name, or text.

browser_file_upload

Upload files through a file input by ref or selector.

browser_paste

Paste clipboard-style text and/or files into a target by snapshot ref, selector, or role/name. Supports text/plain, text/html, text/rtf, and local file paths such as images. If no target is provided, the focused editable element is used, then the root document body.

browser_handle_dialog

Accept or dismiss the current JavaScript dialog.

browser_resize

Resize the browser viewport.

browser_list_devices

List all Playwright device presets available to browser_toggle_device. Returns each preset's viewport, screen, device scale factor, touch/mobile flags, and default browser type. Desktop is included as the restore target.

browser_toggle_device

Toggle the current browser page between desktop and mobile device emulation without recreating the browser context. Uses Playwright device descriptors and Chromium CDP for dynamic mobile metrics, touch, device scale factor, orientation, and user agent. On Firefox/WebKit only viewport resizing is ava…

browser_screenshot

Take a PNG screenshot of the current page.

browser_take_screenshot

Take a PNG screenshot of the current page.

browser_evaluate

Evaluate JavaScript in iframe#rbi-frame.

browser_run_code

Run an async JavaScript snippet with Playwright page and iframe#rbi-frame helpers.

browser_close

Close the current Playwright browser session.

// known CVEs in dependencies4 high

Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.

high@modelcontextprotocol/sdk@1.13.1GHSA-345p-7cg4-v4c7

@modelcontextprotocol/sdk has cross-client data leak via shared server/transport instance reuse

high@modelcontextprotocol/sdk@1.13.1GHSA-8r9q-7v3j-jr4g

Anthropic's MCP TypeScript SDK has a ReDoS vulnerability

high@modelcontextprotocol/sdk@1.13.1GHSA-w48q-cv73-mx4w

Model Context Protocol (MCP) TypeScript SDK does not enable DNS rebinding protection by default

highplaywright@1.52.0GHSA-7mvr-c777-76hp

Playwright downloads and installs browsers without verifying the authenticity of the SSL certificate

Depend on this server? Get alerted when its CVEs change.Watch this server free →

// environment variables

To run this server yourself, you supply these values. They go in your own MCP client configuration and stay on your machine. The secret label means the value is sensitive, not that the server mishandles it.

configSA2_ACTION_TIMEOUT_MS

configSA2_AUTO_LOGIN_BEFORE_NAVIGATE"": "true",

configSA2_BROWSER

configSA2_HEADLESS

configSA2_IGNORE_HTTPS_ERRORS"": "true",

configSA2_LOGIN_REDIRECT_PATH"": "/app/login",

configSA2_LOGIN_REDIRECT_WAIT_MS"": "8000",

configSA2_LOGIN_SETTLE_MS"": "1500",

configSA2_LOGIN_URL"": "https://www.test7878.com/agent/login?clientId=...&clientSecret=...",

configSA2_LOG_FILE"": "/tmp/sa2-browser.log"

configSA2_LOG_LEVEL"": "info",

configSA2_LOG_STDERR"": "true",

configSA2_NAVIGATION_TIMEOUT_MS

configSA2_RBI_FRAME_CONTENT_WAIT_MS"": "15000",

configSA2_RBI_FRAME_SELECTOR"": "#rbi-frame",

configSA2_RBI_FRAME_WAIT_MS"": "5000",

configSA2_TEST_SNAPSHOT_FILTER

configSA2_VIEWPORT_HEIGHT

configSA2_VIEWPORT_WIDTH

// quality suggestions

Tool annotations

No tools have read-only/destructive annotations

Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.

All four hints declared on every tool

40/40 tools missing one or more hints — sa2_help (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); sa2_list_available_targets (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); sa2_open_target (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +37 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.

For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.

Tool inputs are validated

29/40 tool handlers declare input schemas (73%)

Declare an inputSchema with zod/joi/yup on every tool definition.

Tool handlers catch errors

37/40 tool handlers wrap calls in try/catch (93%)

Wrap each tool handler body in try/catch and return a structured error response.

Tests exist

No test files found

Add tests that exercise each declared tool.

No eval / new Function

2 eval() or new Function() calls — dynamic code execution

Replace eval / Function with explicit parsing or safer alternatives.

Production dependencies are patched

0 critical, 4 high severity in production deps — @modelcontextprotocol/sdk@1.13.1 (high), @modelcontextprotocol/sdk@1.13.1 (high)

Run npm audit fix, or upgrade the affected packages to a non-vulnerable version.

Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.

// full audit trail

The findings above are the summary. The full trail, every check we ran, each deduction, the network hosts observed and the dependency advisories, goes to verified publishers, along with an alert whenever a new one lands. Verified publishers can also review each finding and dispute it in one click. Publisher corrections have sharpened several of our checks this month, because the maintainer knows the codebase better than any scanner.

// improvement guidance — verified publishers only

We have 7 concrete improvements we can share with the publisher of this MCP. Each comes with specific guidance to raise the trust score.

// embed badge in your README

[![M8ven Verified](https://m8ven.ai/badge/mcp/sa2web/sa2web-mcp?variant=verified)](https://m8ven.ai/mcp/sa2web/sa2web-mcp)

Shows verification status without the grade. Want the grade badge instead? Remove ?variant=verified from the URL.

commit: 67823d42f43d7f07a48b03da38d1c6a29e10cff5

code hash: 869c435b1a5d1cb1132fe7d8e4452e26e15d5628c157a50a127d8ab77d4da699

view raw JSON →

Check MCPs from inside your assistant

Tool Check · MCP

Vetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.

https://m8ven.ai/api/mcp/tool-check

check_toolsearch_toolscompare_toolsrecommend_alternativescheck_publisherreport_concern

How to add it →Free · no account needed · works in any MCP client