// SCAN REPORT

InSite

52 tools. 30 can modify or destroy data without limits.

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

Last updated: 30 Jun 2026

30 can modify or destroy data

22 read-only

52 tools total

Community server · catalogue entry verified 30/06/2026

How to control InSite ↓

TOOL MAP

What InSite exposes to your agents

Read (22) Write / Execute (28) Destructive / Financial (2)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous InSite tools

Destructive · Low clear_console_logs Clearing stored console logs removes data that cannot be recovered once deleted. Destructive · Low clear_network_logs Clearing logs is irreversible — once network request logs are deleted, they cannot be recovered. Execute · Critical evaluate_js JavaScript execution in a browser context is a classic Execute primitive that can interact with the DOM, fetch data, exfiltrate credentials, modify page state, trigger requests, or perform any action the page can perform Execute · High evaluate_js_on_element This tool allows running arbitrary JavaScript in the browser context, which can perform any operation accessible from JavaScript (DOM manipulation, cookie access, network requests, form submission, etc.).

30 of InSite's 52 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control InSite

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

Deny destructive operations

{
  "clear_console_logs": {
    "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

{
  "close_browser": {
    "limits": [
      {
        "counter": "close_browser_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "capture_full_scrollable_page": {
    "limits": [
      {
        "counter": "capture_full_scrollable_page_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 InSite — 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 INSITE →

Instant setup, no code required.

FULL CATALOGUE

All 52 InSite tools

DESTRUCTIVE 2 tools

Destructive clear_console_logs Clear stored console logs Destructive clear_network_logs Clear stored network request logs

EXECUTE 25 tools

Execute click_element Click an element identified by CSS selector Execute debug_mode Enable or disable comprehensive debugging mode with enhanced logging Execute evaluate_js Execute JavaScript code in the page context Execute evaluate_js_on_element Execute JavaScript code on a specific element Execute go_back Navigate back in browser history Execute go_forward Navigate forward in browser history Execute handle_csp Configure Content Security Policy handling and compliance checking Execute highlight_element Visually highlight elements on the page for debugging purposes Execute hover_element Hover over an element to trigger hover effects Execute jest_adapter Integration adapter for Jest testing framework with browser automation capabilities Execute load_page Navigate to a URL and wait for the page to load Execute manage_certificates Certificate validation and custom certificate handling Execute mocha_adapter Integration adapter for Mocha testing framework with browser automation hooks Execute performance_monitoring Real-time performance monitoring, profiling, and optimization insights Execute playwright_test_adapter Integration adapter for Playwright Test runner with advanced configuration and reporting Execute press_key Press a keyboard key or key combination Execute reload_page Reload the current page Execute scroll_page Scroll the page or a specific element Execute set_geolocation Set the browser geolocation Execute set_viewport_size Set the browser viewport size Execute switch_browser Switch to a different browser engine (chromium, firefox, webkit) Execute trace_execution Enable detailed execution tracing for debugging browser operations Execute validate_security Security validation and vulnerability scanning for web pages Execute wait_for_element Wait for an element to be in a specific state Execute wait_for_navigation Wait for page navigation to complete

WRITE 3 tools

Write close_browser Close the browser and cleanup resources Write set_user_agent Set the browser user agent string Write type_text Type text into an input element

READ 22 tools

Read capture_full_scrollable_page Capture the entire scrollable page height in one long image Read capture_performance_timeline Capture detailed browser performance timeline data for analysis Read cross_browser_visual_validation Validate visual consistency across different browser engines Read element_screenshot_compare Compare specific elements across different states or versions Read error_tracking Advanced error tracking, categorization, and alerting system Read get_browser_info Get information about the current browser engine Read get_console_logs Get console logs from the browser Read get_current_url Get the current page URL Read get_dom Get DOM content as HTML string, optionally filtered by CSS selector Read get_element_info Get comprehensive information about an element using JavaScript Read get_network_logs Get network request logs from the browser Read get_page_title Get the current page title Read get_viewport_info Get current viewport dimensions and device pixel ratio Read list_available_browsers Get list of all available browser engines Read screenshot Take a screenshot of the current page Read screenshot_compare Compare screenshots with diff generation for visual testing Read scroll_to_element_and_screenshot Scroll to a specific element and take a screenshot of it Read session_analytics Session-based analytics, user journey tracking, and workflow optimization Read test_reporter Advanced test reporting and result aggregation with multi-format output and analytics Read usage_analytics Track and report browser automation usage patterns and statistics Read visual_regression_testing Automated visual regression testing with baseline management Read visual_test_reporting Generate comprehensive visual test reports with image galleries

EXPLORE

Related servers

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

FAQ

Questions about InSite

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

Yes. The InSite server exposes 2 destructive tools including clear_console_logs, clear_network_logs. 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 InSite? +

The InSite server has 3 write tools including close_browser, set_user_agent, type_text. 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 InSite.

How many tools does the InSite MCP server expose? +

52 tools across 4 categories: Destructive, Execute, Read, Write. 22 are read-only. 30 can modify, create, or delete data.

How do I enforce a policy on InSite? +

Register the InSite 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 InSite tool call.

Deterministic rules across all 52 InSite tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON INSITE →

Instant setup, no code required.

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