AI agents call diff_to_text to retrieve information from Docx without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The name 'diff_to_text' implies generating a textual representation of differences, which is a read-only operation. However, the description is empty, lowering confidence. In the context of a Word document MCP server focused on track changes and OOXML manipulation, this tool likely reads document differences and returns them as text. No destructive or write behavior is implied by the name alone.
From the tool's definition Tool name 'diff_to_text' suggests converting a diff to text output, which is a read/transform operation with no side effects.
Documented attack patterns abuse exactly the kind of access diff_to_text gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Docx, and nothing reaches the server without passing your rules. This is the rule we recommend for diff_to_text:
{
"version": "1",
"default": "deny",
"tools": {
"diff_to_text": {}
}
}diff_to_text is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
diff_to_text. It is categorised as a Read tool in the Docx MCP Server, which means it retrieves data without modifying state.
Register the Docx MCP server in PolicyLayer and add a rule for diff_to_text: 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 Docx. Nothing to install.
diff_to_text 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 diff_to_text 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 diff_to_text. 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.
diff_to_text is provided by the Docx MCP server (securityronin/docx-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Docx, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
219 Docx tools catalogued and risk-classified — across an index of 43,000+ MCP servers.