AI agents call download_polyhaven_asset to retrieve information from Blender without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Downloading an asset from a public library is a read operation that retrieves data without modifying local systems or executing arbitrary code. No side effects beyond local file storage. Severity is low as Poly Haven is a legitimate public source and downloaded assets are inert until used in Blender.
From the tool's definition Tool name 'download_polyhaven_asset' indicates retrieval of assets from Poly Haven, a public 3D asset library. Description is empty, limiting direct evidence, but the verb 'download' and context of retrieving external assets align with data retrieval.
Attacks that exploit this kind of access
download_polyhaven_asset. It is categorised as a Read tool in the Blender MCP Server, which means it retrieves data without modifying state.
Register the Blender MCP server in PolicyLayer and add a rule for download_polyhaven_asset: 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 Blender. Nothing to install.
download_polyhaven_asset 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 download_polyhaven_asset 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 download_polyhaven_asset. 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.
download_polyhaven_asset is provided by the Blender MCP server (silwings1986/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 →