Merge multiple memories into one. Combines content and tags, then deletes originals.
AI agents call merge_memories to permanently remove resources in Claude Memory MCP Server — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
While the merge creates a new combined memory (a Write action), it irreversibly deletes the original memories as part of the same operation. Per the severity rules, Destructive takes precedence over Write. If misused, an agent could merge and thus destroy multiple important memories without the ability to recover them (unless a backup exists), giving this a high severity blast radius.
From the tool's definition "Combines content and tags, then deletes originals" — the originals are permanently deleted as part of the merge operation.
Attacks that exploit this kind of access
Merge multiple memories into one. Combines content and tags, then deletes originals. It is categorised as a Destructive tool in the Claude Memory MCP Server MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Claude Memory MCP Server MCP server in PolicyLayer and add a rule for merge_memories: 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 Claude Memory MCP Server. Nothing to install.
merge_memories is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the merge_memories 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 merge_memories. 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.
merge_memories is provided by the Claude Memory MCP Server MCP server (seongcheoljeon/claudememorymcp). 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 →