AI agents call immich_search_metadata to retrieve information from Immich without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves or queries data from the photo library based on specified metadata criteria. It has no side effects—it does not create, modify, delete, or execute operations. It is a straightforward search/filter function that returns matching assets without altering the underlying data or triggering external operations.
From the tool's definition Tool performs search operations using metadata filters (date range, type, location, camera model, person, filename) with no indication of modification or deletion capabilities.
Attacks that exploit this kind of access
Search assets using metadata filters: date range, type, location, camera model, person, filename. It is categorised as a Read tool in the Immich MCP Server, which means it retrieves data without modifying state.
Register the Immich MCP server in PolicyLayer and add a rule for immich_search_metadata: 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_search_metadata 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 immich_search_metadata 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_search_metadata. 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_search_metadata 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 →