AI agents invoke run_workflow to trigger actions in ComfyUI MCP Server. 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 runs workflows that generate AI-created content. While not destructive or financial, it executes external operations whose effects depend on workflow arguments—which agent could mis-specify to generate unwanted, resource-intensive, or sensitive content.
From the tool's definition Tool named 'run_workflow' on a server that 'Generate and refine AI images/audio/video through natural conversation.' The tool executes workflows which trigger generation of media content, and given the sibling tools include cancel_job and get_job, this…
Documented attack patterns abuse exactly the kind of access run_workflow gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and ComfyUI MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for run_workflow:
{
"version": "1",
"default": "deny",
"tools": {
"run_workflow": {
"limits": [
{
"counter": "run_workflow_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}run_workflow stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
run_workflow. It is categorised as a Execute tool in the ComfyUI MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the ComfyUI MCP Server MCP server in PolicyLayer and add a rule for run_workflow: 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 ComfyUI MCP Server. Nothing to install.
run_workflow 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 run_workflow 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 run_workflow. 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.
run_workflow is provided by the ComfyUI MCP Server MCP server (joenorton/comfyui-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from ComfyUI MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
15 ComfyUI MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.