AI agents use xhs_save_example to create or update resources in XHS-MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your XHS-MCP environment.
The tool creates or writes files to storage, fitting the Write category. Severity is medium because file creation has limited blast radius—files can be deleted and overwritten—but could consume resources or clutter storage if misused at scale. Confidence is high based on explicit mention of 'save' (write operation) and file format specification.
From the tool's definition Tool name 'xhs_save_example' and description indicating it 'saves example text files (.txt format, filename must end with .txt)' describes creating/storing files, which is reversible data modification.
Attacks that exploit this kind of access
保存范文(txt格式,文件名必须以.txt结尾). It is categorised as a Write tool in the XHS-MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the XHS- MCP server in PolicyLayer and add a rule for xhs_save_example: 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 XHS-MCP. Nothing to install.
xhs_save_example 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 xhs_save_example 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 xhs_save_example. 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.
xhs_save_example is provided by the XHS- MCP server (lastkimi/xhs-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 →