AI agents use apply_filter to create or update resources in Graphics — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Graphics environment.
Based on the naming pattern and server context, apply_filter likely applies a visual filter to an image, modifying it (reversibly in principle, though it may overwrite the original file). This fits the Write category. Confidence is reduced because the description is empty, leaving the exact behavior (in-place overwrite vs. new file creation) unknown. Severity is medium given potential to overwrite image files.
From the tool's definition Tool name 'apply_filter' on a graphics manipulation server with sibling tools like convert_image, crop_image, flip_image, resize_image, rotate_image — all of which modify images in place or produce modified outputs.
Attacks that exploit this kind of access
apply_filter. It is categorised as a Write tool in the Graphics MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Graphics MCP server in PolicyLayer and add a rule for apply_filter: 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 Graphics. Nothing to install.
apply_filter 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 apply_filter 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 apply_filter. 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.
apply_filter is provided by the Graphics MCP server (lesleslie/graphics-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 →