AI agents call tool_get_schema to retrieve information from Mcp Excel without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves schema metadata (column names and types) from an Excel/CSV file without executing queries, modifying data, or triggering external operations. It is a read-only introspection operation with minimal blast radius. An AI agent misusing this tool cannot cause data loss, financial impact, or unintended code execution.
From the tool's definition Tool description states it 'Get[s] column names and types for a table or view' — a retrieval operation with no modification or execution of code. Name and description both indicate pure data inspection.
Attacks that exploit this kind of access
Get column names and types for a table or view. It is categorised as a Read tool in the Mcp Excel MCP Server, which means it retrieves data without modifying state.
Register the Mcp Excel MCP server in PolicyLayer and add a rule for tool_get_schema: 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 Mcp Excel. Nothing to install.
tool_get_schema 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 tool_get_schema 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 tool_get_schema. 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.
tool_get_schema is provided by the Mcp Excel MCP server (ivan-loh/mcp-excel). 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 →