AI agents call search_assets to retrieve information from Rapid7 without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves asset/endpoint information from a SIEM (InsightIDR), which is sensitive security infrastructure data. While categorically a Read operation (no data modification), the severity is high because unauthorized asset enumeration in a SIEM could enable reconnaissance for lateral movement, privilege escalation, or targeting of specific systems.
From the tool's definition search_assets queries InsightIDR endpoints by hostname, IP address, OS, or agent status—a pure retrieval operation with no modifications or deletions. The description explicitly states 'search' and 'by' parameters with no side effects mentioned.
Attacks that exploit this kind of access
Search InsightIDR assets (endpoints) by hostname, IP address, OS, or agent status. It is categorised as a Read tool in the Rapid7 MCP Server, which means it retrieves data without modifying state.
Register the Rapid7 MCP server in PolicyLayer and add a rule for search_assets: 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 Rapid7. Nothing to install.
search_assets 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_assets 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_assets. 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_assets is provided by the Rapid7 MCP server (solomonneas/rapid7-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 →