AI agents call memory_get to retrieve information from Memory without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves data by ID and only increments a hit counter (a minor side effect used for metrics). There is no creation, modification, deletion, or execution of arbitrary operations. The side effect is purely observational and does not affect the substance of the memory data itself. This is a straightforward Read operation.
From the tool's definition Tool name 'memory_get' and description 'Retrieve a single memory entry by ID. Increments the hit counter.' indicate a read-only retrieval operation with no data modification or deletion.
Attacks that exploit this kind of access
Retrieve a single memory entry by ID. Increments the hit counter. It is categorised as a Read tool in the Memory MCP Server, which means it retrieves data without modifying state.
Register the Memory MCP server in PolicyLayer and add a rule for memory_get: 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 Memory. Nothing to install.
memory_get 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 memory_get 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_get. 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_get is provided by the Memory MCP server (joshdougall/memory-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 →