delete_file
AI agents call delete_file to permanently remove resources in Filesystem MCP Server — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
Deletion of files is a destructive operation that irreversibly removes data. Even with directory sandboxing in place, an AI agent misusing this tool could permanently erase important files within the sandbox. This warrants the Destructive category and high severity due to the potential for data loss, though blast radius is somewhat mitigated by the documented directory restrictions.
From the tool's definition Tool named 'delete_file' with no description, but the name unambiguously indicates irreversible deletion of data. The server description acknowledges 'file reading/writing, directory management' but this tool performs an operation that cannot be undone.
Attacks that exploit this kind of access
delete_file. It is categorised as a Destructive tool in the Filesystem MCP Server MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Filesystem MCP Server MCP server in PolicyLayer and add a rule for delete_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 Filesystem MCP Server. Nothing to install.
delete_file 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 delete_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 delete_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.
delete_file is provided by the Filesystem MCP Server MCP server (preston-harrison/fs-mcp-py). 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 →