// SCAN REPORT

Scout

32 tools. 21 can modify or destroy data without limits.

1 destructive tool with no built-in limits. Policy required.

Last updated: 30 Jun 2026

21 can modify or destroy data

11 read-only

32 tools total

Community server · catalogue entry verified 30/06/2026

How to control Scout ↓

TOOL MAP

What Scout exposes to your agents

Read (11) Write / Execute (20) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Scout tools

Destructive · High scout_delete_post This tool permanently removes content from a platform. Execute · Medium scout_handoff_cancel This tool triggers an external operation in the browser — cancelling a pending handoff and removing UI elements. Execute · High scout_evaluate This tool allows execution of arbitrary JavaScript code within a live browser page, which is a hallmark of the Execute category. Execute · High scout_navigate While the tool itself is relatively constrained (it navigates and returns metadata), it falls into Execute rather than Read because: (1) navigation is an action that changes browser state and can trigger side effects on

21 of Scout's 32 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Scout

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

Deny destructive operations

{
  "scout_delete_post": {
    "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

{
  "scout_post": {
    "limits": [
      {
        "counter": "scout_post_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "scout_block_resources": {
    "limits": [
      {
        "counter": "scout_block_resources_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 Scout — 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 SCOUT →

Instant setup, no code required.

FULL CATALOGUE

All 32 Scout tools

DESTRUCTIVE 1 tools

Destructive scout_delete_post Delete one of your posts on a platform.

EXECUTE 17 tools

Execute scout_back Go back in browser history. Execute scout_click Click an element by its snapshot ID. Returns healer result describing what changed. If stateChange is Execute scout_drag Drag one element onto another. Execute scout_evaluate Execute JavaScript in the current page and return the result. Use for: reading page state, clicking elements b Execute scout_forward Go forward in browser history. Execute scout_handoff Inject a banner in the live browser asking the user to take a manual action. Returns IMMEDIATELY with a handof Execute scout_handoff_cancel Cancel a pending handoff and remove the banner from the browser. Execute scout_hover Hover over an element by its snapshot ID. Execute scout_login Log in to a social platform automatically. Handles multi-step flows and unusual activity challenges. Auto-save Execute scout_navigate Navigate to a URL. Returns {url, title} by default (lean). Set snapshot=true for full accessibility tree + scr Execute scout_new_tab Open a new browser tab, optionally navigating to a URL. Execute scout_press_key Press a keyboard key (e.g. Enter, Escape, ArrowDown, Backspace). Execute scout_refresh Refresh the current page. Execute scout_scroll Scroll the page or a specific element (modal, dialog) in a direction. Use element_id to scroll within a contai Execute scout_select Select an option from a dropdown/select element by its snapshot ID. Execute scout_switch_tab Switch to a browser tab by index (from scout_tabs). Execute scout_wait Wait for a page condition.

WRITE 3 tools

Write scout_post Compose and publish a post on a social platform (linkedin, x, medium). Uses the platform adapter pattern — han Write scout_save_session Save the current browser session (cookies, localStorage) to a named file. Write scout_type Type text into an input element by its snapshot ID. Uses React-safe keyboard events. If stateChange is

READ 11 tools

Read scout_block_resources Block certain resource types to speed up loading (e.g. image, media, stylesheet, font, script). Read scout_console_logs Get the last 100 browser console logs and errors. Read scout_elements Get the current page Read scout_handoff_check Check whether a pending handoff has been completed by the human. Returns immediately. Call this every 5-10 sec Read scout_list_sessions List all saved browser sessions. Read scout_load_session Load a saved browser session. This will restart the browser context. Read scout_my_posts Get your recent posts on a platform. Read scout_screenshot Take a plain screenshot of the current page without element badges. Read scout_search_posts Search for posts on a platform. Read scout_snapshot Take a snapshot of the current page. Returns accessibility tree (numbered elements). Set lite=true to skip the Read scout_tabs List all open browser tabs with their index, URL, title, and which is active.

EXPLORE

Related servers

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

FAQ

Questions about Scout

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

Yes. The Scout server exposes 1 destructive tools including scout_delete_post. 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 Scout? +

The Scout server has 3 write tools including scout_post, scout_save_session, scout_type. 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 Scout.

How many tools does the Scout MCP server expose? +

32 tools across 4 categories: Destructive, Execute, Read, Write. 11 are read-only. 21 can modify, create, or delete data.

How do I enforce a policy on Scout? +

Register the Scout 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 Scout tool call.

Deterministic rules across all 32 Scout tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON SCOUT →

Instant setup, no code required.

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