搜索网页并抓取前几个结果的内容
AI agents call web_search_and_scrape to retrieve information from Web Search MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and extracts publicly available web content. It has no side effects on any system or data store. The operations are read-only: searching and scraping return information without altering state. Severity is low because misuse would only allow an agent to gather existing public information, posing minimal direct harm compared to write, execute, or destructive operations.
From the tool's definition Tool name 'web_search_and_scrape' and description indicating it 'searches webpages and scrapes the content of top results' — performs information retrieval only without creating, modifying, or deleting data.
Attacks that exploit this kind of access
搜索网页并抓取前几个结果的内容. It is categorised as a Read tool in the Web Search MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Web Search MCP Server MCP server in PolicyLayer and add a rule for web_search_and_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 Web Search MCP Server. Nothing to install.
web_search_and_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 web_search_and_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 web_search_and_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.
web_search_and_scrape is provided by the Web Search MCP Server MCP server (mantraa-zzz/web_search). 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 →