AI agents call inspect_query to retrieve information from Duckdb without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool appears designed to examine or analyze queries (likely their structure, plan, or metadata) rather than execute them or modify data. Inspecting a query is a non-destructive, side-effect-free operation that retrieves information about a query, placing it squarely in the Read category.
From the tool's definition Tool name 'inspect_query' and description 'Inspect a query in the DuckDB database' indicate read-only inspection of query details without execution or modification. The term 'inspect' denotes examination rather than execution, creation, or deletion.
Attacks that exploit this kind of access
Inspect a query in the DuckDB database. It is categorised as a Read tool in the Duckdb MCP Server, which means it retrieves data without modifying state.
Register the Duckdb MCP server in PolicyLayer and add a rule for inspect_query: 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 Duckdb. Nothing to install.
inspect_query 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 inspect_query 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 inspect_query. 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.
inspect_query is provided by the Duckdb MCP server (synehq/mcp-server-duckdb). 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 →