AI agents call search_design_system to retrieve information from GDS MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves or queries design system metadata through keyword search. It does not create, modify, delete, or execute external operations. It is purely informational and read-only, consistent with other sibling tools like list_tokens, list_components, and get_token. The low severity reflects minimal risk from misuse—search queries cannot damage or alter the design system data.
From the tool's definition Tool description indicates it performs keyword search across design system metadata (color tokens, components, guidelines). The Korean description translates to 'Search for color tokens, components, and guidelines by keyword all at once.
Attacks that exploit this kind of access
컬러 토큰, 컴포넌트, 가이드라인을 한 번에 키워드 검색합니다. 어떤 도구를 써야 할지 모를 때 먼저 사용하세요. It is categorised as a Read tool in the GDS MCP MCP Server, which means it retrieves data without modifying state.
Register the GDS MCP server in PolicyLayer and add a rule for search_design_system: 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 GDS MCP. Nothing to install.
search_design_system 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_design_system 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_design_system. 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_design_system is provided by the GDS MCP server (supin-kim-ilevit-com/gds-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 →