AI agents call drupal_query_jsonapi to retrieve information from Drupal without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool explicitly performs only GET requests against the JSON:API, making it a read-only operation. However, it allows arbitrary endpoint access including custom resources and index endpoints, which could expose sensitive data across the entire Drupal site. The 'escape hatch' nature means it can reach any readable resource, raising the severity to medium due to potential broad data exposure.
From the tool's definition arbitrary GET against the JSON:API
Attacks that exploit this kind of access
Escape hatch: arbitrary GET against the JSON:API. Use when the higher-level tools do not cover what you need (custom resources, /jsonapi/index, etc.). It is categorised as a Read tool in the Drupal MCP Server, which means it retrieves data without modifying state.
Register the Drupal MCP server in PolicyLayer and add a rule for drupal_query_jsonapi: 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 Drupal. Nothing to install.
drupal_query_jsonapi 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 drupal_query_jsonapi 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 drupal_query_jsonapi. 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.
drupal_query_jsonapi is provided by the Drupal MCP server (lucaspretti/drupal-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 →