AI agents invoke runServerDirectly to trigger actions in MCP Server Generator. 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.
Running a server directly is an Execute action—it triggers external operations (server startup/process execution) whose effects depend on which server is targeted. This carries high severity as it could spawn arbitrary processes or services. The confidence is not higher due to the empty tool description, but the name and server purpose provide strong contextual evidence.
From the tool's definition Tool name 'runServerDirectly' indicates direct execution of server processes. Combined with server context of 'Creates, manages, and registers custom MCP servers', this tool appears to execute arbitrary server code.
Documented attack patterns abuse exactly the kind of access runServerDirectly gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MCP Server Generator, and nothing reaches the server without passing your rules. This is the rule we recommend for runServerDirectly:
{
"version": "1",
"default": "deny",
"tools": {
"runServerDirectly": {
"limits": [
{
"counter": "runserverdirectly_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}runServerDirectly 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.
runServerDirectly. It is categorised as a Execute tool in the MCP Server Generator MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the MCP Server Generator MCP server in PolicyLayer and add a rule for runServerDirectly: 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 MCP Server Generator. Nothing to install.
runServerDirectly 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 runServerDirectly 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 runServerDirectly. 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.
runServerDirectly is provided by the MCP Server Generator MCP server (serhatuzbas/mcp-server-generator). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from MCP Server Generator, 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.
79 MCP Server Generator tools catalogued and risk-classified — across an index of 43,000+ MCP servers.