Crop clip.
AI agents use vfx_crop to create or update resources in vidMagik-mcp — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your vidMagik-mcp environment.
Cropping modifies video data in a new output (reversible write operation). While non-destructive in intent, it transforms media content and could produce unintended results if applied incorrectly by an AI agent (e.g., cropping away critical content). Severity is medium because the operation is reversible and the blast radius is limited to the output video artifact, not permanent data loss or external systems.
From the tool's definition Tool name 'vfx_crop' and description 'Crop clip' indicate modification of video content. Cropping is a reversible transformation that alters video dimensions/framing without deleting the source file.
Attacks that exploit this kind of access
Crop clip. It is categorised as a Write tool in the vidMagik-mcp MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the vidMagik- MCP server in PolicyLayer and add a rule for vfx_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 vidMagik-mcp. Nothing to install.
vfx_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 vfx_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 vfx_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.
vfx_crop is provided by the vidMagik- MCP server (vizionik25/vidmagik-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 →