Get the schema/structure of a specific table
AI agents call get_schema to retrieve information from Vanna AI MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a read-only query operation to retrieve schema metadata. It has no side effects, cannot modify or delete data, and cannot execute arbitrary code. The blast radius of misuse is minimal — an attacker could learn table structures but cannot exfiltrate sensitive data from tables themselves, execute queries, or cause damage.
From the tool's definition Tool name is 'get_schema' and description states 'Get the schema/structure of a specific table' — retrieves metadata about database structure without modification, deletion, or execution of operations.
Attacks that exploit this kind of access
Get the schema/structure of a specific table. It is categorised as a Read tool in the Vanna AI MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Vanna AI MCP Server MCP server in PolicyLayer and add a rule for 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 Vanna AI MCP Server. Nothing to install.
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 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 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.
get_schema is provided by the Vanna AI MCP Server MCP server (lilgreml1n/vanna-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 →