AI agents call download_file to retrieve information from Nutstore without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves/downloads a file from cloud storage to a local machine. It has no side effects on the source data—it does not create, modify, delete, or overwrite anything in the cloud storage or trigger external operations. This is a straightforward read operation. Severity is low because downloading files poses minimal risk even if misused by an AI agent.
From the tool's definition Tool name 'download_file' and description '从坚果云下载单个文件到本机本地路径' (download a single file from Nutstore to local path) indicates retrieval of data without modification or deletion of source data.
Attacks that exploit this kind of access
从坚果云下载单个文件到本机本地路径。. It is categorised as a Read tool in the Nutstore MCP Server, which means it retrieves data without modifying state.
Register the Nutstore MCP server in PolicyLayer and add a rule for download_file: 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 Nutstore. Nothing to install.
download_file 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 download_file 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 download_file. 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.
download_file is provided by the Nutstore MCP server (silverze/nutstore-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 →