AI agents call lookup_safety_ratings to retrieve information from VIN MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs data retrieval only. It queries public NHTSA safety rating information by VIN and returns existing safety metrics. There are no side effects, no data modification, no code execution, and no destructive or financial operations. The blast radius of misuse is minimal—an attacker could only access publicly available vehicle safety information, which poses no security or privacy risk.
From the tool's definition Tool 'lookup_safety_ratings' retrieves NHTSA crash test data and returns ratings without modifying any data. Description explicitly states it 'Get[s]' safety ratings and 'Returns' test scores—purely informational query operations.
Attacks that exploit this kind of access
Get NHTSA crash test safety ratings for a vehicle by VIN. Returns overall rating and individual test scores (frontal, side, rollover). It is categorised as a Read tool in the VIN MCP MCP Server, which means it retrieves data without modifying state.
Register the VIN MCP server in PolicyLayer and add a rule for lookup_safety_ratings: 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 VIN MCP. Nothing to install.
lookup_safety_ratings 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 lookup_safety_ratings 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 lookup_safety_ratings. 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.
lookup_safety_ratings is provided by the VIN MCP server (keptlive/vin-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 →