AI agents call sg_cpi_latest to retrieve information from Sgdata without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and returns publicly available Singapore government CPI statistics. It performs a read-only operation with no side effects, no data modification, and no irreversible actions. The blast radius of misuse is minimal since it only exposes aggregate economic statistics that are already public.
From the tool's definition Tool name 'sg_cpi_latest' and description 'Most recent CPI index value and YoY change for a matching' indicate retrieval of CPI (Consumer Price Index) data. The verb 'get' is implicit in 'latest' which retrieves the most recent published value.
Attacks that exploit this kind of access
Most recent CPI index value and YoY change for a matching. It is categorised as a Read tool in the Sgdata MCP Server, which means it retrieves data without modifying state.
Register the Sgdata MCP server in PolicyLayer and add a rule for sg_cpi_latest: 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 Sgdata. Nothing to install.
sg_cpi_latest 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 sg_cpi_latest 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 sg_cpi_latest. 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.
sg_cpi_latest is provided by the Sgdata MCP server (sypherin/sgdata-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 →