AI agents call brave_search_summary to retrieve information from MCP2Brave without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs web search queries and returns summarized results. It retrieves and queries data from the web without any side effects, data modification, or code execution. It falls squarely into the Read category as a search/query operation.
From the tool's definition Tool description states '使用Brave搜索引擎搜索网络信息' (use Brave search engine to search web information). The name 'brave_search_summary' and sibling tools (get_url_content_direct, search_brave_with_summary, search_news, url_content) all indicate read-only web search…
Attacks that exploit this kind of access
使用Brave搜索引擎搜索网络信息. It is categorised as a Read tool in the MCP2Brave MCP Server, which means it retrieves data without modifying state.
Register the MCP2Brave MCP server in PolicyLayer and add a rule for brave_search_summary: 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 MCP2Brave. Nothing to install.
brave_search_summary 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 brave_search_summary 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 brave_search_summary. 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.
brave_search_summary is provided by the MCP2Brave MCP server (mcp2everything/mcp2brave). 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 →