preview_raw
AI agents call preview_raw to retrieve information from Rawtherapee without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool name and context indicate it generates or retrieves a preview of RAW image content for display purposes. No modification, deletion, or execution of external commands is implied. Although the description is empty, the sibling tools (add_, adjust_, apply_) all modify state, whereas preview_ is read-only. This is a retrieval/query operation with no side effects, fitting the Read category.
From the tool's definition Tool name 'preview_raw' suggests viewing or displaying RAW image data without modification. The server description emphasizes a 'visual feedback loop' for iterative editing, and this tool appears to support that feedback mechanism by retrieving preview data.
Attacks that exploit this kind of access
preview_raw. It is categorised as a Read tool in the Rawtherapee MCP Server, which means it retrieves data without modifying state.
Register the Rawtherapee MCP server in PolicyLayer and add a rule for preview_raw: 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 Rawtherapee. Nothing to install.
preview_raw 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 preview_raw 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 preview_raw. 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.
preview_raw is provided by the Rawtherapee MCP server (lucamarien/rawtherapee-mcp-server). 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 →