Use an AI agent specialized in performing computer actions safely and effectively.
AI agents invoke computer_action_agent to trigger actions in OpenAI Agents 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.
The tool is explicitly designed to perform computer actions via an AI agent. Computer actions include clicking, typing, running applications, browser automation, and other interactions with a computer system. This constitutes executing operations on a system whose effects depend on arguments.
From the tool's definition performing computer actions
Documented attack patterns abuse exactly the kind of access computer_action_agent gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and OpenAI Agents MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for computer_action_agent:
{
"version": "1",
"default": "deny",
"tools": {
"computer_action_agent": {
"limits": [
{
"counter": "computer_action_agent_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}computer_action_agent 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.
Use an AI agent specialized in performing computer actions safely and effectively. It is categorised as a Execute tool in the OpenAI Agents MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the OpenAI Agents MCP Server MCP server in PolicyLayer and add a rule for computer_action_agent: 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 OpenAI Agents MCP Server. Nothing to install.
computer_action_agent 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 computer_action_agent 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 computer_action_agent. 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.
computer_action_agent is provided by the OpenAI Agents MCP Server MCP server (lroolle/openai-agents-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from OpenAI Agents 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.
4 OpenAI Agents MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.