immich_duplicates_delete
AI agents call immich_duplicates_delete to permanently remove resources in Immich — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
The tool name explicitly includes 'delete', which removes data irreversibly. Even though duplicates might be considered less critical than unique assets, deletion of any photo data cannot be undone and represents a Destructive action category. High severity due to potential for unintended photo loss if an AI agent misuses this tool or misidentifies duplicates.
From the tool's definition Tool name 'immich_duplicates_delete' indicates irreversible deletion operation. Server description mentions 'safe deletion workflows' but the empty tool description prevents confirming safeguards.
Attacks that exploit this kind of access
immich_duplicates_delete. It is categorised as a Destructive tool in the Immich MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Immich MCP server in PolicyLayer and add a rule for immich_duplicates_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 Immich. Nothing to install.
immich_duplicates_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 immich_duplicates_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 immich_duplicates_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.
immich_duplicates_delete is provided by the Immich MCP server (whitehara/immich-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 →