Capture current viewport as base64 image. Takes screenshot of current 3D viewport view. Limited to 800px max dimension for performance. Args: - max_size (optional): Maximum dimension in pixels (100-800, default: 800) Returns: Base64 encoded PNG image data with metadata Use when: Visualizing scene...
AI agents call blender_get_screenshot to retrieve information from ClaudeKit Blender MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool only retrieves visual data from the current 3D scene state without altering anything. It has no side effects on the scene, materials, objects, or any other system state. The use case 'Visualizing scene state, checking results, debugging' confirms it is purely observational.
From the tool's definition Tool captures and returns a screenshot of the current viewport as base64 image data. The description states 'Capture current viewport as base64 image' and the return value is 'Base64 encoded PNG image data with metadata'.
Attacks that exploit this kind of access
Capture current viewport as base64 image. Takes screenshot of current 3D viewport view. Limited to 800px max dimension for performance. Args: - max_size (optional): Maximum dimension in pixels (100-800, default: 800) Returns: Base64 encoded PNG image data with metadata Use when: Visualizing scene state, checking results, debugging Don. It is categorised as a Read tool in the ClaudeKit Blender MCP MCP Server, which means it retrieves data without modifying state.
Register the ClaudeKit Blender MCP server in PolicyLayer and add a rule for blender_get_screenshot: 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 ClaudeKit Blender MCP. Nothing to install.
blender_get_screenshot is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the blender_get_screenshot 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 blender_get_screenshot. 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.
blender_get_screenshot is provided by the ClaudeKit Blender MCP server (olbboy/claudekit-blender-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 →