AI agents use merge_cells to create or update resources in Excel — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Excel environment.
Merging cells is a reversible modification operation that alters worksheet structure and formatting. It does not retrieve data (Read), execute code (Execute), irreversibly destroy data (Destructive), or involve financial transactions (Financial). It fits squarely in Write: creates or modifies data reversibly.
From the tool's definition Tool name is 'merge_cells' and description states 'Merge a range of cells'. The operation modifies cell structure and formatting within a worksheet but does not delete data or create irreversible changes (cells can be unmerged).
Attacks that exploit this kind of access
Merge a range of cells. It is categorised as a Write tool in the Excel MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Excel MCP server in PolicyLayer and add a rule for merge_cells: 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 Excel. Nothing to install.
merge_cells 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 merge_cells 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 merge_cells. 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.
merge_cells is provided by the Excel MCP server (shmaxi/excel-mcp-server-node). 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 →