set_model · Nanobanana MCP: Risk & Policy

WHAT IT DOES

What set_model does on Nanobanana

AI agents use set_model to create or update resources in Nanobanana — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Nanobanana environment.

WHY IT NEEDS A POLICY

Why set_model needs a policy

This tool creates or modifies session settings (the active Gemini model) in a reversible manner. It does not delete data, execute external code, move money, or perform irreversible operations. It is a configuration change that can be undone by calling the tool again with a different model. Classified as Write rather than Read because it mutates session state rather than retrieving data.

From the tool's definition Tool name 'set_model' and description 'Set the Gemini model for this session' indicate modification of session configuration state.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

set_model Rate-limited

set_model stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "set_model": {
      "limits": [
        {
          "counter": "set_model_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

LIMIT THIS TOOL →

Instant setup, no code required.

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

FAQ

Questions about set_model

What does the set_model tool do? +

Set the Gemini model for this session. It is categorised as a Write tool in the Nanobanana MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on set_model? +

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

What risk level is set_model? +

set_model is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit set_model? +

Yes. Add a rate_limit block to the set_model 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 set_model completely? +

Set action: deny in the PolicyLayer policy for set_model. 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 set_model? +

set_model is provided by the Nanobanana MCP server (@ycse/nanobanana-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.