AI agents use qf_export_csv to create or update resources in Qingflow — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Qingflow environment.
This tool exports data to a persistent CSV file, which is a form of data creation and persistence. While file export itself is less severe than modifying application data, it creates artifacts and could be misused to exfiltrate or stage data. It's reversible (files can be deleted), distinguishing it from Destructive operations.
From the tool's definition Tool creates a new CSV file by exporting query results, which is a reversible write operation. Description states 'Export...to a CSV file and return file path', indicating file creation/writing.
Attacks that exploit this kind of access
Export list query result to a CSV file and return file path + summary instead of large inline payloads. It is categorised as a Write tool in the Qingflow MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Qingflow MCP server in PolicyLayer and add a rule for qf_export_csv: 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 Qingflow. Nothing to install.
qf_export_csv 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 qf_export_csv 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 qf_export_csv. 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.
qf_export_csv is provided by the Qingflow MCP server (qingflow-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 →