AI agents use crop-dicom-image to create or update resources in DICOM-MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your DICOM-MCP environment.
Cropping a DICOM image modifies the image data by removing boundary regions. This is a Write operation as it transforms the loaded image. However, without clarity on whether it overwrites the original or creates a new output, there is some ambiguity. Medical image modification carries medium severity due to potential clinical implications if original data is altered.
From the tool's definition 'Crop a loaded DICOM image by removing boundary percentage' — modifies image data by cropping/removing portions
Attacks that exploit this kind of access
Crop a loaded DICOM image by removing boundary percentage. It is categorised as a Write tool in the DICOM-MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the DICOM- MCP server in PolicyLayer and add a rule for crop-dicom-image: 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 DICOM-MCP. Nothing to install.
crop-dicom-image is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the crop-dicom-image 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 crop-dicom-image. 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.
crop-dicom-image is provided by the DICOM- MCP server (shaunporwal/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 →