AI agents use copy_blob to create or update resources in Gcp Storage — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Gcp Storage environment.
Copying a blob is a write operation that creates new data in the destination location. While reversible (the copy can be deleted), it modifies storage state and could consume quota/costs. It does not delete or destroy data (so not Destructive), does not execute arbitrary code (not Execute), and does not involve financial transactions (not Financial).
From the tool's definition Tool is named 'copy_blob' and described as 'Copy a blob from one location to another.' This creates a new copy of data within GCS, modifying the state of storage by adding a new object.
Documented attack patterns abuse exactly the kind of access copy_blob gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Gcp Storage, and nothing reaches the server without passing your rules. This is the rule we recommend for copy_blob:
{
"version": "1",
"default": "deny",
"tools": {
"copy_blob": {
"limits": [
{
"counter": "copy_blob_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}copy_blob stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Copy a blob from one location to another. It is categorised as a Write tool in the Gcp Storage MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Gcp Storage MCP server in PolicyLayer and add a rule for copy_blob: 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 Gcp Storage. Nothing to install.
copy_blob 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_blob 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_blob. 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_blob is provided by the Gcp Storage MCP server (uysalserkan/gcp-storage-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Gcp Storage, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
23 Gcp Storage tools catalogued and risk-classified — across an index of 43,000+ MCP servers.