Brief description: Deletes files generated in OUTPUT_PATH (or default temporary directory) and performs basic resource cleanup. Call only when the user explicitly indicates deletion of temporary or output files.
AI agents call cleanup_resources to permanently remove resources in Math MCP Server — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
The tool explicitly deletes files, which is an irreversible destructive action. It targets output files and temporary directories, potentially wiping multiple generated artifacts. The high severity reflects the blast radius — misuse could delete all output files from a session. Confidence is high because the description clearly states 'Deletes files'.
From the tool's definition 'Deletes files generated in OUTPUT_PATH (or default temporary directory) and performs basic resource cleanup'
Documented attack patterns abuse exactly the kind of access cleanup_resources gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Math MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for cleanup_resources:
{
"version": "1",
"default": "deny",
"hide": [
"cleanup_resources"
]
}cleanup_resources disappears from the agent's tool list entirely, and any attempt to call it is denied. The rest of the server keeps working.
Free to start. No card required.
Brief description: Deletes files generated in OUTPUT_PATH (or default temporary directory) and performs basic resource cleanup. Call only when the user explicitly indicates deletion of temporary or output files. It is categorised as a Destructive tool in the Math MCP Server MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Math MCP Server MCP server in PolicyLayer and add a rule for cleanup_resources: 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 Math MCP Server. Nothing to install.
cleanup_resources is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the cleanup_resources 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 cleanup_resources. 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.
cleanup_resources is provided by the Math MCP Server MCP server (111-test-111/math-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Math MCP Server, 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.
22 Math MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.