Pause machine motion immediately (feed hold). Sends
AI agents invoke feed_hold to trigger actions in Cnc Fluidnc. 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.
This tool sends a real-time command to a CNC machine to immediately pause motion. It is an Execute-category action because it triggers an external physical operation (halting machine movement). Severity is high because misuse on an actively cutting CNC router could cause material damage, tool breakage, or unsafe machine states, though it is safer than commands that start or resume motion.
From the tool's definition 'Pause machine motion immediately (feed hold)' - triggers an immediate external operation on physical CNC hardware
Attacks that exploit this kind of access
Pause machine motion immediately (feed hold). Sends. It is categorised as a Execute tool in the Cnc Fluidnc MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Cnc Fluidnc MCP server in PolicyLayer and add a rule for feed_hold: 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 Cnc Fluidnc. Nothing to install.
feed_hold 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 feed_hold 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 feed_hold. 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.
feed_hold is provided by the Cnc Fluidnc MCP server (whitneydesignlabs/cnc-fluidnc-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 →