AI agents use uploadFile to create or update resources in SMART-E2B — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your SMART-E2B environment.
uploadFile creates new files or modifies existing file state in the sandbox through file upload. This is a Write operation rather than Read (which only retrieves data) or Execute (which runs code). While uploads could theoretically introduce malicious content, the sandboxed nature limits blast radius to the isolated environment, justifying 'medium' severity.
From the tool's definition Tool enables 'Upload a file to the sandbox' - a reversible creation/modification operation that writes data to the sandboxed environment.
Attacks that exploit this kind of access
Upload a file to the sandbox. It is categorised as a Write tool in the SMART-E2B MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the SMART-E2B MCP server in PolicyLayer and add a rule for uploadFile: 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 SMART-E2B. Nothing to install.
uploadFile is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the uploadFile 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 uploadFile. 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.
uploadFile is provided by the SMART-E2B MCP server (leghis/smart-e2b). 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 →