Read any web page as clean text + markdown. Give a URL, get back the page title, readable
AI agents call scrape to retrieve information from Superhighway without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool fetches and parses web content for consumption; it has no side effects, does not modify data, and does not execute code or commands. The verb 'Read' and the passive nature of web scraping (extracting published content) align with the Read category. Severity is low because retrieval of public web content poses minimal risk even if misused by an agent.
From the tool's definition Tool description states 'Read any web page as clean text + markdown' — explicitly a read operation that retrieves and returns page content without modification.
Attacks that exploit this kind of access
Read any web page as clean text + markdown. Give a URL, get back the page title, readable. It is categorised as a Read tool in the Superhighway MCP Server, which means it retrieves data without modifying state.
Register the Superhighway MCP server in PolicyLayer and add a rule for scrape: 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 Superhighway. Nothing to install.
scrape is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the scrape 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 scrape. 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.
scrape is provided by the Superhighway MCP server (patwalls/superhighway-mcp). 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 →