Use when: you have a single photo
AI agents call get-photo to retrieve information from Apple Photos without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves photo metadata or content from a local macOS photo library. It has no side effects—it only reads data. Even if it returns photo content (EXIF data, image file, etc.), that is still a Read operation. The blast radius of misuse is minimal: an AI could retrieve unintended photos, potentially exposing personal images, but cannot modify, delete, or execute code.
From the tool's definition Tool name 'get-photo' and description indicate retrieval of a single photo from the Apple Photos library. The server description states it 'enables AI assistants to query and export', positioning get-photo as a query operation.
Attacks that exploit this kind of access
Use when: you have a single photo. It is categorised as a Read tool in the Apple Photos MCP Server, which means it retrieves data without modifying state.
Register the Apple Photos MCP server in PolicyLayer and add a rule for get-photo: 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 Apple Photos. Nothing to install.
get-photo 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 get-photo 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 get-photo. 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.
get-photo is provided by the Apple Photos MCP server (sweetrb/apple-photos-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 →