Read a CSV file and return per-column statistics:
AI agents call csv_stats to retrieve information from Claude MCP Starter without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool only retrieves and analyzes data from an existing CSV file without creating, modifying, deleting, or executing any code. The operation is purely computational analysis of data already present, resulting in a Read classification. Severity is low because misuse would only expose data that's already accessible via the read operation, with no side effects or destructive potential.
From the tool's definition Tool performs a read operation on CSV files and returns statistics (count, mean, median, etc.). The description explicitly states 'Read a CSV file and return per-column statistics' with no modification, deletion, or execution capabilities.
Attacks that exploit this kind of access
Read a CSV file and return per-column statistics:. It is categorised as a Read tool in the Claude MCP Starter MCP Server, which means it retrieves data without modifying state.
Register the Claude MCP Starter MCP server in PolicyLayer and add a rule for csv_stats: 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 Claude MCP Starter. Nothing to install.
csv_stats 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 csv_stats 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 csv_stats. 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.
csv_stats is provided by the Claude MCP Starter MCP server (justdreameritis/claude-mcp-starter). 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 →