AI agents call extract_previews to retrieve information from Darktable without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and transforms image data into previews (a read operation with rendering side effects that do not persist back to the source). It does not create, modify, delete, or execute arbitrary operations on the photo library itself. Auto-rotation is a viewing transformation, not a destructive edit.
From the tool's definition Tool name 'extract_previews' and description 'Extract auto-rotated JPEG previews from a directory' indicate data retrieval with no modification of source files.
Attacks that exploit this kind of access
Extract auto-rotated JPEG previews from a directory of. It is categorised as a Read tool in the Darktable MCP Server, which means it retrieves data without modifying state.
Register the Darktable MCP server in PolicyLayer and add a rule for extract_previews: 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 Darktable. Nothing to install.
extract_previews 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 extract_previews 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 extract_previews. 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.
extract_previews is provided by the Darktable MCP server (w1ne/darktable-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 →