MCP tool that launch search with Brave
AI agents call search_web_with_brave to retrieve information from Tagny MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries the Brave search engine and returns results. It is a read-only operation consistent with the server's described purpose of 'web search using Brave and DuckDuckGo engines.' No data is created, modified, or deleted. Severity is low as the worst case is retrieval of unintended web content. Confidence is slightly reduced because the description is minimal and doesn't detail parameters or output scope.
From the tool's definition 'launch search with Brave' — performs a web search, retrieving data from the web without side effects
Attacks that exploit this kind of access
MCP tool that launch search with Brave. It is categorised as a Read tool in the Tagny MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Tagny MCP Server MCP server in PolicyLayer and add a rule for search_web_with_brave: 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 Tagny MCP Server. Nothing to install.
search_web_with_brave 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 search_web_with_brave 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 search_web_with_brave. 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.
search_web_with_brave is provided by the Tagny MCP Server MCP server (tagny/tagny-mcp-server). 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 →