AI agents call get_download_url to retrieve information from Yadisk without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a query/retrieval operation that returns data (a URL) without modifying, executing code, deleting, or moving money. It is purely informational and has no capability to modify Yandex Disk state. The temporary nature of the URL further confirms it is a safe read operation.
From the tool's definition The tool 'get_download_url' retrieves a temporary download URL for a file. The description explicitly states it 'Get[s]' a URL, which is a read-only retrieval operation with no side effects. The URL is temporary, meaning it does not persist indefinitely.
Attacks that exploit this kind of access
Get a temporary direct download URL for a file. It is categorised as a Read tool in the Yadisk MCP Server, which means it retrieves data without modifying state.
Register the Yadisk MCP server in PolicyLayer and add a rule for get_download_url: 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 Yadisk. Nothing to install.
get_download_url 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_download_url 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_download_url. 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_download_url is provided by the Yadisk MCP server (patr56/yadisk-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 →