Crop an image locally to a target aspect or requested dimensions.
AI agents use auto_crop to create or update resources in Jgkme/kilo Image Gen — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Jgkme/kilo Image Gen environment.
Cropping an image is a reversible write operation that modifies image content and dimensions. While the changes can theoretically be undone (original could be re-processed or recovered), the tool actively transforms image data. This is less severe than destructive operations (original file not necessarily deleted), but more impactful than simple reads.
From the tool's definition Tool description states 'Crop an image locally to a target aspect or requested dimensions' - this modifies image data by removing portions and resizing dimensions.
Attacks that exploit this kind of access
Crop an image locally to a target aspect or requested dimensions. It is categorised as a Write tool in the Jgkme/kilo Image Gen MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Jgkme/kilo Image Gen MCP server in PolicyLayer and add a rule for auto_crop: 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 Jgkme/kilo Image Gen. Nothing to install.
auto_crop 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 auto_crop 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 auto_crop. 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.
auto_crop is provided by the Jgkme/kilo Image Gen MCP server (jgkme/img-gen-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 →