Execute JavaScript in the current page and return the result. Use for: reading page state, clicking elements blocked by shadow DOM overlays, interacting with Web Components, or any DOM operation not covered by other tools. The code runs in the page context with full DOM access. Returns the serial...
AI agents invoke scout_evaluate to trigger actions in Scout. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool allows execution of arbitrary JavaScript code within a live browser page, which is a hallmark of the Execute category. While it does not directly delete data or move funds, it can perform any DOM operation, trigger arbitrary actions, exfiltrate data, modify page state, or interact with page-based APIs.
From the tool's definition Tool description explicitly states 'Execute JavaScript in the current page' and 'runs in the page context with full DOM access.' This grants arbitrary code execution within the browser page context.
Attacks that exploit this kind of access
Execute JavaScript in the current page and return the result. Use for: reading page state, clicking elements blocked by shadow DOM overlays, interacting with Web Components, or any DOM operation not covered by other tools. The code runs in the page context with full DOM access. Returns the serialized result. It is categorised as a Execute tool in the Scout MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Scout MCP server in PolicyLayer and add a rule for scout_evaluate: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Scout. Nothing to install.
scout_evaluate is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the scout_evaluate rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for scout_evaluate. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
scout_evaluate is provided by the Scout MCP server (lautrek/scout). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →