AI agents invoke create_tool to trigger actions in Open WebUI 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 creates a new custom tool containing arbitrary Python code, which can be executed within the Open WebUI platform. Registering executable code is effectively an Execute-class action with high blast radius — a malicious or buggy tool definition could run arbitrary Python in the server context. Write would be too lenient given the code-execution nature of the artifact being created.
From the tool's definition Create a new custom tool with Python code
Documented attack patterns abuse exactly the kind of access create_tool gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Open WebUI MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for create_tool:
{
"version": "1",
"default": "deny",
"tools": {
"create_tool": {
"limits": [
{
"counter": "create_tool_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}create_tool 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.
Create a new custom tool with Python code. It is categorised as a Execute tool in the Open WebUI MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Open WebUI MCP Server MCP server in PolicyLayer and add a rule for create_tool: 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 Open WebUI MCP Server. Nothing to install.
create_tool 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 create_tool 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 create_tool. 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.
create_tool is provided by the Open WebUI MCP Server MCP server (troylar/open-webui-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Open WebUI 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.
82 Open WebUI MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.