AI agents call suggest_sql_from_schema to retrieve information from Xugu without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool only validates and suggests improvements to SQL; it does not execute queries, modify data, or perform any side effects. This is a read-only analytical operation. While the server context mentions 'full DDL/DML support', this specific tool is explicitly limited to validation and suggestions in lightweight mode, making it a low-severity Read operation.
From the tool's definition Tool description states 'Validate and suggest improvements for user-provided SQL (lightweight mode)' - this is a validation and suggestion operation with no execution or modification of data.
Attacks that exploit this kind of access
Validate and suggest improvements for user-provided SQL (lightweight mode). It is categorised as a Read tool in the Xugu MCP Server, which means it retrieves data without modifying state.
Register the Xugu MCP server in PolicyLayer and add a rule for suggest_sql_from_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 Xugu. Nothing to install.
suggest_sql_from_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 suggest_sql_from_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 suggest_sql_from_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.
suggest_sql_from_schema is provided by the Xugu MCP server (leokupo/xugu-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 →