AI agents invoke load_level to trigger actions in O3de. 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.
Loading/opening a level in the O3DE Editor is an external operation that changes the editor's active context. It's not a simple read (it triggers editor actions), not destructive (it doesn't delete data), not financial, but it does execute an editor command with side effects on the running application state, warranting the Execute category.
From the tool's definition 'Open a level in the O3DE Editor' — triggers an external operation in the editor application, changing its active state
Documented attack patterns abuse exactly the kind of access load_level gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and O3de, and nothing reaches the server without passing your rules. This is the rule we recommend for load_level:
{
"version": "1",
"default": "deny",
"tools": {
"load_level": {
"limits": [
{
"counter": "load_level_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}load_level 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.
Open a level in the O3DE Editor. It is categorised as a Execute tool in the O3de MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the O3de MCP server in PolicyLayer and add a rule for load_level: 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 O3de. Nothing to install.
load_level 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 load_level 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 load_level. 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.
load_level is provided by the O3de MCP server (nickschuetz/o3de-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from O3de, 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.
30 O3de tools catalogued and risk-classified — across an index of 43,000+ MCP servers.