Soft-delete a memory by ID (tombstone). The memory is marked as deleted but remains in storage and can be recovered. Use memory_purge for permanent deletion.
AI agents call memory_delete to permanently remove resources in Copilot Memory Store — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
Although the deletion is soft (tombstone) and technically recoverable, it is still a delete operation that removes a memory from active use. The description explicitly contrasts it with memory_purge for permanent deletion, indicating this is the less severe variant. However, delete operations are classified as Destructive by category.
From the tool's definition 'Soft-delete a memory by ID (tombstone)' and 'The memory is marked as deleted but remains in storage and can be recovered'
Documented attack patterns abuse exactly the kind of access memory_delete gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Copilot Memory Store, and nothing reaches the server without passing your rules. This is the rule we recommend for memory_delete:
{
"version": "1",
"default": "deny",
"hide": [
"memory_delete"
]
}memory_delete disappears from the agent's tool list entirely, and any attempt to call it is denied. The rest of the server keeps working.
Free to start. No card required.
Soft-delete a memory by ID (tombstone). The memory is marked as deleted but remains in storage and can be recovered. Use memory_purge for permanent deletion. It is categorised as a Destructive tool in the Copilot Memory Store MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Copilot Memory Store MCP server in PolicyLayer and add a rule for memory_delete: 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 Copilot Memory Store. Nothing to install.
memory_delete 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 memory_delete 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_delete. 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_delete is provided by the Copilot Memory Store MCP server (timothywarner-org/copilot-memory-store). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Copilot Memory Store, 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.
7 Copilot Memory Store tools catalogued and risk-classified — across an index of 43,000+ MCP servers.