process_raw
AI agents invoke process_raw to trigger actions in Rawtherapee. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
The server description explicitly mentions running RawTherapee CLI to process RAW photos. A tool named 'process_raw' almost certainly triggers CLI execution to develop/render RAW image files. This falls under Execute as it runs an external process. Severity is high due to the potential for processing arbitrary files via CLI.
From the tool's definition Tool name 'process_raw' on a server described as enabling 'RAW photo development via RawTherapee CLI' — CLI execution is central to the server's operation.
Attacks that exploit this kind of access
process_raw. It is categorised as a Execute tool in the Rawtherapee MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Rawtherapee MCP server in PolicyLayer and add a rule for process_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.
process_raw is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the process_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 process_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.
process_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 →