AI agents call list_dir to retrieve information from MCP Files without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries filesystem structure within a constrained sandbox environment. It has no side effects, does not create, modify, delete, or execute anything. The sandboxing and access controls further limit blast radius. It is a straightforward Read operation with minimal risk.
From the tool's definition Tool description explicitly states it 'List files/directories' and 'hides `.metadata` and paths outside the sandbox', indicating a read-only operation that retrieves directory contents without modification.
Risk signalsAdmin/system-level operation
Attacks that exploit this kind of access
List files/directories under the workspace root. Accepts glob pattern, recursion, and metadata toggles; hides .metadata and paths outside the sandbox. It is categorised as a Read tool in the MCP Files MCP Server, which means it retrieves data without modifying state.
Register the MCP Files MCP server in PolicyLayer and add a rule for list_dir: 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 MCP Files. Nothing to install.
list_dir 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 list_dir 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 list_dir. 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.
list_dir is provided by the MCP Files MCP server (tspspi/mcpfiles). 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 →