AI agents use copy_from_primary to create or update resources in I18n — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your I18n environment.
This tool creates or modifies translation data across multiple locales without deletion. While it could introduce incorrect translations if misused, the effect is reversible (can be corrected by other Write tools like add_translation or corrected via copy_from_primary again).
From the tool's definition Tool copies primary locale values to other locales, modifying translation file content across specified locales. Description explicitly states it performs copy/assignment operations on translations.
Attacks that exploit this kind of access
Copy the primary locale value verbatim to specified locales for specified keys. It is categorised as a Write tool in the I18n MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the I18n MCP server in PolicyLayer and add a rule for copy_from_primary: 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 I18n. Nothing to install.
copy_from_primary is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the copy_from_primary 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 copy_from_primary. 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.
copy_from_primary is provided by the I18n MCP server (robin-heat/i18n-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 →