AI agents use memory_save_image to create or update resources in Mementos — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Mementos environment.
This tool creates new data (image memories) or modifies existing records in the memory system. It is reversible since the Mementos system provides entity_delete and bulk_forget tools for removal. The severity is medium rather than high because the blast radius of an agent saving unintended images is limited to information storage; the impact depends on what images are stored and later retrieved.
From the tool's definition Tool description states 'Save an image memory' and 'Saves with content_type=', indicating it creates or modifies data by storing an image into the memory system.
Attacks that exploit this kind of access
Save an image memory. If OPENAI_API_KEY is set and image_url provided, auto-extracts a description via GPT-4o-mini vision. Saves with content_type=. It is categorised as a Write tool in the Mementos MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mementos MCP server in PolicyLayer and add a rule for memory_save_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 Mementos. Nothing to install.
memory_save_image is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the memory_save_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 memory_save_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.
memory_save_image is provided by the Mementos MCP server (@hasna/mementos). 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 →