query_patients
AI agents call query_patients to retrieve information from Simple Dicom without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries patient information from medical imaging servers (DICOM/PACS). While the category is Read (no data modification or deletion), the severity is high because DICOM patient data contains Protected Health Information (PHI) subject to HIPAA and similar regulations.
From the tool's definition Tool name 'query_patients' on a DICOM server context (PACS/VNA) that retrieves patient data; sibling tools include 'query_instances', 'query_series', 'query_studies' which are all read operations; no description provided but function is clear from name and…
Attacks that exploit this kind of access
query_patients. It is categorised as a Read tool in the Simple Dicom MCP Server, which means it retrieves data without modifying state.
Register the Simple Dicom MCP server in PolicyLayer and add a rule for query_patients: 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 Simple Dicom. Nothing to install.
query_patients 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 query_patients 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 query_patients. 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.
query_patients is provided by the Simple Dicom MCP server (thalesmms/simple-dicom-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 →