AI agents call get_backup_vault to retrieve information from Storage 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 information about a backup vault without side effects. It falls cleanly into the Read category. Severity is low because the blast radius of inadvertently exposing backup vault metadata is minimal—no data is deleted, modified, or financial transactions are triggered. Confidence is high given the clear read-only semantics.
From the tool's definition Tool name is 'get_backup_vault' and description states 'Gets a backup vault.' The verb 'get' and the read-only nature of retrieving vault metadata indicates no data modification, creation, deletion, or external operation triggering.
Attacks that exploit this kind of access
Gets a backup vault. It is categorised as a Read tool in the Storage MCP Server, which means it retrieves data without modifying state.
Register the Storage MCP server in PolicyLayer and add a rule for get_backup_vault: 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 Storage. Nothing to install.
get_backup_vault 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_backup_vault 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_backup_vault. 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_backup_vault is provided by the Storage MCP server (@google-cloud/storage-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.