AI agents use memory_lock 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 acquires a write lock, which is a reversible state modification operation. While locks can cause temporary contention or deadlock if misused, the lock itself is not destructive (it can be released), not financial, and not arbitrary code execution. It falls under Write because it modifies the state of the memory system by establishing exclusive access.
From the tool's definition The tool description explicitly states it "Acquire[s] an exclusive write lock on a memory key" — this is a concurrency control operation that modifies system state (lock acquisition) and prepares for write operations, but does not itself delete or execute…
Attacks that exploit this kind of access
Acquire an exclusive write lock on a memory key to prevent concurrent writes. 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_lock: 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_lock 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_lock 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_lock. 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_lock 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 →