AI agents use strings_export_repo_sync to create or update resources in Phrase — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Phrase environment.
While export operations might seem read-like, this tool explicitly syncs/writes translations to a repository provider, which constitutes data modification. The async nature and multi-step process increase the blast radius if misused by an agent (e.g., overwriting translations in the wrong repository). However, it's not Destructive because the operation is reversible—translations can be re-synced or reverted.
From the tool's definition Tool performs 'Export translations from Phrase Strings to a repository provider' which involves writing/syncing data to an external repository.
Attacks that exploit this kind of access
Export translations from Phrase Strings to a repository provider according to the .phrase.yml file. Export is asynchronous and may take several seconds. It is categorised as a Write tool in the Phrase MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Phrase MCP server in PolicyLayer and add a rule for strings_export_repo_sync: 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 Phrase. Nothing to install.
strings_export_repo_sync 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 strings_export_repo_sync 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 strings_export_repo_sync. 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.
strings_export_repo_sync is provided by the Phrase MCP server (phrase-mcp-server). 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 →