AI agents use import_project to create or update resources in Clo3d — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Clo3d environment.
Importing a project file alters the current working state of the CLO3D scene (overwriting or merging project data). While not permanently destructive or irreversible on the file system, it changes the in-memory scene state, which can override unsaved work. This is a Write operation—it creates/modifies application state reversibly.
From the tool's definition Tool imports/loads files into the CLO3D scene state via 'Open a CLO project or garment file (.zprj/.zpac/.avt) into the scene', which modifies the active scene by replacing or populating it with design data.
Attacks that exploit this kind of access
Open a CLO project or garment file (.zprj/.zpac/.avt) into the scene. It is categorised as a Write tool in the Clo3d MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Clo3d MCP server in PolicyLayer and add a rule for import_project: 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 Clo3d. Nothing to install.
import_project 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 import_project 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 import_project. 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.
import_project is provided by the Clo3d MCP server (lilbonner/clo3d-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 →