// SCAN REPORT

Wuying AgentBay

59 tools. 41 can modify or destroy data without limits.

4 destructive tools with no built-in limits. Policy required.

Last updated: 12 Jun 2026

41 can modify or destroy data

18 read-only

59 tools total

Community server · catalogue entry verified 12/06/2026

How to control Wuying AgentBay ↓

TOOL MAP

What Wuying AgentBay exposes to your agents

Read (18) Write / Execute (37) Destructive / Financial (4)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Wuying AgentBay tools

Destructive · Medium browser_close Closing a browser page destroys its current state, session, and any unsaved data. Destructive · Medium browser_tab_close Closing a tab cannot be undone; any unsaved data, active sessions, or in-progress work in that tab is permanently lost. Destructive · Medium close_window Closing a window is an irreversible action that terminates the window and any unsaved state within it. Destructive · High release_resource Releasing a cloud resource (session, container, or VM) is effectively irreversible: once the environment is torn down, any in-memory state, temporary files, or running processes are permanently destroyed.

41 of Wuying AgentBay's 59 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Wuying AgentBay

PolicyLayer is an MCP gateway — it sits between your AI agents and Wuying AgentBay, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "browser_close": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "browser_file_upload": {
    "limits": [
      {
        "counter": "browser_file_upload_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "browser_snapshot": {
    "limits": [
      {
        "counter": "browser_snapshot_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Wuying AgentBay — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON WUYING AGENTBAY →

Free to start. No card required.

FULL CATALOGUE

All 59 Wuying AgentBay tools

DESTRUCTIVE 4 tools

Destructive browser_close Close the page Destructive browser_tab_close Close a tab Destructive close_window Close a specific window by its window ID. Destructive release_resource release resource when task finished

EXECUTE 28 tools

Execute browser_navigate Navigate to a URL Execute browser_navigate_back Go back to the previous page Execute browser_navigate_forward Go forward to the next page Execute browser_tab_new Open a new tab Execute browser_wait_for Wait for text to appear or disappear or a specified time to pass Execute start_app Start a specified application using the provided command and optional working directory. Returns a list of pro Execute stop_app_by_cmd Terminate an application using the provided stop command. Use with caution as this will forcefully terminate t Execute stop_app_by_pid Terminate a specific process identified by its Process ID. Use with caution as this will forcefully terminate Execute stop_app_by_pname Stop all processes associated with a specified process name. Use with caution as this will forcefully terminat Execute activate_window Activate a specific window by its window ID. Execute browser_click Perform click on a web page Execute browser_drag Perform drag and drop between two elements Execute browser_generate_playwright_test Generate a Playwright test for given scenario Execute browser_handle_dialog Handle a dialog Execute browser_hover Hover over element on page Execute browser_install Install the browser specified in the config. Call this if you get an error about the browser not being install Execute browser_press_key Press a key on the keyboard Execute browser_resize Resize the browser window Execute browser_select_option Select an option in a dropdown Execute browser_tab_select Select a tab by index Execute browser_type Type text into editable element Execute focus_mode Enable or disable focus mode. When focus mode is enabled, only windows from the currently active process and i Execute fullscreen_window Set a specific window to fullscreen mode by its window ID. Execute maximize_window Maximize a specific window by its window ID. Execute minimize_window Minimize a specific window by its window ID. Execute resize_window Resize a specific window by its window ID. Execute restore_window Restore a specific window to its normal state by its window ID. Execute shell Executes a shell command with timeout and returns the output or an error on linux platform.

WRITE 9 tools

Write browser_file_upload Upload one or multiple files Write browser_pdf_save Save page as PDF Write create_directory Create a new directory or ensure a directory exists. Can create multiple nested directories in one operation. Write edit_file Make line-based edits to a text file. Each edit replaces exact line sequences with new content. Returns a git- Write move_file Move or rename files and directories. Can move files between directories and rename them in a single operation Write oss_env_init Create and initialize OSS environment variables with the specified endpoint, access key ID, access key secret, Write oss_upload Upload a local file or directory to the specified OSS bucket. Note: You must call the oss_env_init tool to ini Write oss_upload_annon Upload a local file or directory to the specified URL using HTTP PUT. If a directory is specified, it will be Write write_file Create a new file or write content to an existing file. You can choose to completely overwrite the file or app

READ 18 tools

Read browser_snapshot Capture accessibility snapshot of the current page, this is better than screenshot Read browser_tab_list List browser tabs Read browser_take_screenshot Take a screenshot of the current page. You can Read get_active_window Retrieve information about the currently active window. Returns details including window ID, title, process ID Read get_file_info Retrieve detailed metadata about a file or directory. Returns comprehensive information including size, creati Read get_installed_apps Retrieve a list of installed applications on the system. Supports filtering by Start Menu entries and Desktop Read get_resource The command to retrieve a wuying mcp runtime URL when user wants to get access to this runtime. Each retrieved Read list_directory Get a detailed listing of all files and directories in a specified path. Results clearly distinguish between f Read list_root_windows List all root windows with their associated information. Returns a list of root windows, including their windo Read list_visible_apps List all applications with visible windows, including their associated process information. Returns a list of Read oss_download Download an object from the specified OSS bucket to the given local path. Note: You must call the oss_env_init Read oss_download_annon Download a file from the specified URL to the given local path. If the parent directory does not exist, it wil Read read_file Read the contents of a file from the file system. You can specify an optional Read read_multiple_files Read the contents of multiple files simultaneously. This is more efficient than reading files one by one when Read search_files Recursively search for files and directories matching a pattern. Searches through all subdirectories from the Read system_screenshot Captures a full-screen screenshot of the current display and returns a shareable URL. The screenshot is automa Read browser_console_messages Returns all console messages Read browser_network_requests Returns all network requests since loading the page

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Wuying AgentBay

Can an AI agent delete data through the Wuying AgentBay MCP server? +

Yes. The Wuying AgentBay server exposes 4 destructive tools including browser_close, browser_tab_close, close_window. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Wuying AgentBay? +

The Wuying AgentBay server has 9 write tools including browser_file_upload, browser_pdf_save, create_directory. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Wuying AgentBay.

How many tools does the Wuying AgentBay MCP server expose? +

59 tools across 3 categories: Execute, Read, Write. 18 are read-only. 41 can modify, create, or delete data.

How do I enforce a policy on Wuying AgentBay? +

Register the Wuying AgentBay MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Wuying AgentBay tool call.

Deterministic rules across all 59 Wuying AgentBay tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON WUYING AGENTBAY →

Free to start. No card required.

59 Wuying AgentBay tools catalogued and risk-classified — across an index of 43,000+ MCP servers.