vlm_generate_preview · Albumentations MCP: Risk & Policy

WHAT IT DOES

What vlm_generate_preview does on Albumentations

AI agents invoke vlm_generate_preview to trigger actions in Albumentations. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.

WHY IT NEEDS A POLICY

Why vlm_generate_preview needs a policy

This tool invokes an external generative model (VLM) to create images from text prompts, which constitutes triggering an external operation rather than a simple read or write. The output depends on the arguments supplied and involves computational execution of a generative pipeline. Severity is medium as misuse could generate inappropriate content or consume significant compute resources, but blast radius is limited.

From the tool's definition "Generate an image from a text prompt" — triggers an external VLM (Vision Language Model) operation to produce image output from arbitrary text input

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

vlm_generate_preview Rate-limited

vlm_generate_preview stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "vlm_generate_preview": {
      "limits": [
        {
          "counter": "vlm_generate_preview_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

RATE-LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches Albumentations, so you can run agents against it safely.

FAQ

Questions about vlm_generate_preview

What does the vlm_generate_preview tool do? +

Generate an image from a text prompt for quick ideation. It is categorised as a Execute tool in the Albumentations MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on vlm_generate_preview? +

Register the Albumentations MCP server in PolicyLayer and add a rule for vlm_generate_preview: 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 Albumentations. Nothing to install.

What risk level is vlm_generate_preview? +

vlm_generate_preview is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit vlm_generate_preview? +

Yes. Add a rate_limit block to the vlm_generate_preview 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.

How do I block vlm_generate_preview completely? +

Set action: deny in the PolicyLayer policy for vlm_generate_preview. 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.

What MCP server provides vlm_generate_preview? +

vlm_generate_preview is provided by the Albumentations MCP server (ramsi-k/albumentations-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.