AI agents call kanji to retrieve information from Yomitan without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Even though kanji only reads data, uncontrolled read access leaks sensitive information and racks up API costs — an agent caught in a retry loop can make thousands of calls a minute without anyone noticing.
Attacks that exploit this kind of access
漢字の詳細情報（読み、画数、意味等）を検索する. It is categorised as a Read tool in the Yomitan MCP Server, which means it retrieves data without modifying state.
Register the Yomitan MCP server in PolicyLayer and add a rule for kanji: 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 Yomitan. Nothing to install.
kanji 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 kanji 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 kanji. 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.
kanji is provided by the Yomitan MCP server (yomitan-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.