AI agents use ftp_upload to create or update resources in Ftp — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Ftp environment.
ftp_upload creates or modifies files on a remote FTP/SFTP server. While reversible (unlike ftp_delete), it can overwrite existing data and deploy potentially malicious content to production systems, making it a Write-category risk. Severity is high because an AI agent given unconstrained upload access could compromise server integrity, inject malware, or overwrite critical application files.
From the tool's definition Tool name 'ftp_upload' and description 'Standard single-file transport to the remote server' indicate file creation/modification on a remote system.
Attacks that exploit this kind of access
Standard single-file transport to the remote server. For bulk transfers, favor ftp_batch_upload. It is categorised as a Write tool in the Ftp MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Ftp MCP server in PolicyLayer and add a rule for ftp_upload: 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 Ftp. Nothing to install.
ftp_upload 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 ftp_upload 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 ftp_upload. 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.
ftp_upload is provided by the Ftp MCP server (kynlos/ftp-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 →