Edit one or more images using Google Gemini. Provide images and instructions for how to modify them. Returns a base64-encoded image.
AI agents invoke edit_image to trigger actions in Nano Banana Pro MCP. 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.
The tool invokes an external AI model (Google Gemini) to modify images based on user-supplied instructions. This is an Execute-category action because it runs an external operation whose effects depend on the arguments provided. It is not purely Write (no persistent storage implied) nor Destructive (original images are not necessarily deleted).
From the tool's definition 'Edit one or more images using Google Gemini' and 'instructions for how to modify them' — triggers an external operation (Google Gemini API call) that processes and transforms images based on arbitrary instructions.
Documented attack patterns abuse exactly the kind of access edit_image gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Nano Banana Pro MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for edit_image:
{
"version": "1",
"default": "deny",
"tools": {
"edit_image": {
"limits": [
{
"counter": "edit_image_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}edit_image 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.
Free to start. No card required.
Edit one or more images using Google Gemini. Provide images and instructions for how to modify them. Returns a base64-encoded image. It is categorised as a Execute tool in the Nano Banana Pro MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Nano Banana Pro MCP server in PolicyLayer and add a rule for edit_image: 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 Nano Banana Pro MCP. Nothing to install.
edit_image is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the edit_image 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 edit_image. 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.
edit_image is provided by the Nano Banana Pro MCP server (mrafaeldie12/nano-banana-pro-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Nano Banana Pro MCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
3 Nano Banana Pro MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.