// SCAN REPORT

Pydoll

57 tools. 38 can modify or destroy data without limits.

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

Last updated: 11 Jun 2026

38 can modify or destroy data

19 read-only

57 tools total

Community server · catalogue entry verified 11/06/2026

How to control Pydoll ↓

TOOL MAP

What Pydoll exposes to your agents

Read (19) Write / Execute (36) Destructive / Financial (2)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Pydoll tools

Destructive · Medium clear_cache Clearing cache and storage is an irreversible deletion of data (cached files, localStorage, sessionStorage, cookies, etc.). Destructive · Medium close_tab Closing a browser tab is a destructive action in the sense that any unsaved state, active session, or in-progress work within that tab is permanently lost and cannot be recovered. Execute · High execute_automation_script This tool triggers execution of browser automation scripts whose effects depend entirely on the script content—clicking elements, navigating pages, entering data, submitting forms. Execute · Critical execute_javascript JavaScript execution in a browser context permits arbitrary computation and side effects dependent entirely on the supplied code argument.

38 of Pydoll's 57 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Pydoll

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

Deny destructive operations

{
  "clear_cache": {
    "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

{
  "export_to_csv": {
    "limits": [
      {
        "counter": "export_to_csv_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "analyze_content_with_ai": {
    "limits": [
      {
        "counter": "analyze_content_with_ai_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 Pydoll — 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 PYDOLL →

Free to start. No card required.

FULL CATALOGUE

All 57 Pydoll tools

DESTRUCTIVE 2 tools

Destructive clear_cache Clear browser cache and storage Destructive close_tab Close a specific tab in a browser

EXECUTE 28 tools

Execute execute_automation_script Execute predefined automation scripts Execute execute_javascript Execute JavaScript code in the browser context Execute navigate_to Navigate to a specific URL in a browser tab Execute new_tab Create a new tab in a browser instance Execute start_browser Start a new browser instance with specified configuration Execute stop_browser Stop a browser instance and clean up resources Execute block_requests Block specific network requests Execute bypass_cloudflare Attempt to bypass Cloudflare Turnstile protection Execute bypass_recaptcha Attempt to bypass reCAPTCHA v3 protection Execute download_file Download a file from a URL or trigger a download Execute evade_detection Apply comprehensive evasion techniques Execute go_back Navigate back in browser history Execute inject_script_library Inject JavaScript libraries into the page Execute intelligent_search Intelligently perform search on any website with automatic element detection Execute randomize_fingerprint Randomize browser fingerprint to avoid tracking Execute refresh_page Refresh the current page in a browser tab Execute rotate_proxy Rotate to a new proxy server Execute simulate_human_behavior Simulate realistic human interaction patterns Execute spoof_headers Spoof HTTP headers to appear more legitimate Execute throttle_network Simulate different network conditions Execute click_element Click on a web element with human-like behavior Execute enable_stealth_mode Enable advanced stealth mode to avoid bot detection Execute handle_bot_detection Handle generic bot detection challenges Execute intercept_network_requests Intercept and modify network requests and responses Execute modify_request_headers Modify HTTP request headers Execute randomize_timing Add random delays and timing variations Execute set_active_tab Switch to a specific tab in a browser Execute type_text Type text into an input element with realistic human typing

WRITE 8 tools

Write export_to_csv Export extracted data to CSV format Write export_to_json Export extracted data to JSON format Write generate_pdf Generate a PDF of the current page Write manage_downloads Manage browser downloads (list, pause, resume, cancel) Write save_har Save HTTP Archive (HAR) file of network activity Write save_session Save browser session state for later restoration Write set_user_agent Set a custom or random user agent string Write upload_file Upload a file to a web form

READ 19 tools

Read analyze_content_with_ai Analyze page content using AI for insights and recommendations Read analyze_performance Analyze page performance metrics and provide optimization suggestions Read check_protection_status Check current protection and detection status Read extract_api_responses Extract and save API responses Read extract_data Extract structured data from the current page Read fetch_domain_commands Fetch all possible commands available for the current domain from Chrome DevTools Protocol Read find_element Find a web element using natural attributes or traditional selectors Read get_browser_status Get detailed status information for a specific browser Read get_current_url Get the current URL of a browser tab Read get_network_logs Retrieve detailed network activity logs Read get_page_source Get the HTML source code of the current page Read get_page_title Get the title of the current page Read get_parent_element Get the parent element of a specific element with its attributes Read list_browsers List all active browser instances with their status Read list_tabs List all tabs in a browser instance Read load_session Load a previously saved browser session Read monitor_websockets Monitor WebSocket connections and messages Read take_element_screenshot Take a screenshot of a specific element Read take_screenshot Take a screenshot of the current page or viewport

EXPLORE

Related servers

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

FAQ

Questions about Pydoll

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

Yes. The Pydoll server exposes 2 destructive tools including clear_cache, close_tab. 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 Pydoll? +

The Pydoll server has 8 write tools including export_to_csv, export_to_json, generate_pdf. 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 Pydoll.

How many tools does the Pydoll MCP server expose? +

57 tools across 4 categories: Destructive, Execute, Read, Write. 19 are read-only. 38 can modify, create, or delete data.

How do I enforce a policy on Pydoll? +

Register the Pydoll 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 Pydoll tool call.

Deterministic rules across all 57 Pydoll tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON PYDOLL →

Free to start. No card required.

57 Pydoll tools catalogued and risk-classified — across an index of 43,000+ MCP servers.