browse_categories
AI agents call browse_categories to retrieve information from Korean Stat without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool name indicates a listing/browsing action on categories within a statistics database. No evidence of data modification, deletion, code execution, or financial operations. Without a description, confidence is reduced, but contextual sibling tools and the tool name pattern strongly suggest this is a Read operation that retrieves category information from the KOSIS OpenAPI.
From the tool's definition Tool name 'browse_categories' suggests browsing/listing functionality typical of data retrieval operations. Description is empty, limiting confidence.
Attacks that exploit this kind of access
browse_categories. It is categorised as a Read tool in the Korean Stat MCP Server, which means it retrieves data without modifying state.
Register the Korean Stat MCP server in PolicyLayer and add a rule for browse_categories: 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 Korean Stat. Nothing to install.
browse_categories 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 browse_categories 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 browse_categories. 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.
browse_categories is provided by the Korean Stat MCP server (seolcoding/korean-stat-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 →