AI agents use set_annotation to create or update resources in MCP Figma — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your MCP Figma environment.
This tool falls under Write category because it creates or modifies data (annotations) in a reversible manner. Annotations are metadata attached to design elements and can be edited or removed later. The severity is medium because misuse could clutter designs with unwanted annotations or mislead collaborators, but the impact is limited to annotation metadata rather than core design assets.
From the tool's definition 'Create or update an annotation' – the tool modifies existing annotations or creates new ones, which are reversible changes to design metadata in Figma.
Attacks that exploit this kind of access
Create or update an annotation. It is categorised as a Write tool in the MCP Figma MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the MCP Figma MCP server in PolicyLayer and add a rule for set_annotation: 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 MCP Figma. Nothing to install.
set_annotation 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 set_annotation 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 set_annotation. 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.
set_annotation is provided by the MCP Figma MCP server (yelowflash09/figma_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 →