AI agents call get_icon_info to retrieve information from Spindle without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves icon metadata or properties from the Ameba Design System Spindle without modifying, executing, or deleting anything. It is a simple query operation consistent with other sibling 'get_*' tools on the server (get_accessibility_docs, get_component_info, etc.). The blast radius of misuse is minimal—an AI agent could only retrieve unwanted design information, not cause harm or state changes.
From the tool's definition Tool name is 'get_icon_info' and description translates to 'Get information about a specified icon'. The verb 'get' and the absence of any modification, deletion, or execution semantics indicate a read-only retrieval operation.
Attacks that exploit this kind of access
指定したアイコンの情報を取得します. It is categorised as a Read tool in the Spindle MCP Server, which means it retrieves data without modifying state.
Register the Spindle MCP server in PolicyLayer and add a rule for get_icon_info: 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 Spindle. Nothing to install.
get_icon_info 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 get_icon_info 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 get_icon_info. 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.
get_icon_info is provided by the Spindle MCP server (@openameba/spindle-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.