This tool utilizes OpenAI's model to autonomously execute general-purpose browser-based tasks with balanced performance and reliability using a cloud browser. It handles complex interactions effectively with practical reasoning and clear execution. Optimal for tasks requiring: - Reliable, general...
Risk signalsHigh parameter count (11 properties)
Part of the Hyperbrowser server.
Free to start. No card required.
AI agents may call openai_computer_use_agent to permanently remove or destroy resources in Hyperbrowser. Without a policy, an autonomous agent could delete critical data in a loop with no way to undo the damage. PolicyLayer blocks destructive tools by default and requires explicit human approval before enabling them.
Without a policy, an AI agent could call openai_computer_use_agent in a loop, permanently destroying resources in Hyperbrowser. There is no undo for destructive operations. PolicyLayer blocks this tool by default and only allows it when a human explicitly approves the action.
Destructive tools permanently remove data. Block by default. Only enable with explicit approval workflows.
{
"version": "1",
"default": "deny",
"hide": [
"openai_computer_use_agent"
]
}See the full Hyperbrowser policy for all 10 tools.
These attack patterns abuse exactly the kind of access openai_computer_use_agent gives an agent. Each links to the full case and the policy that stops it:
Other destructive tools across the catalogue. The same approach applies to each: deny by default, or require human approval.
This tool utilizes OpenAI's model to autonomously execute general-purpose browser-based tasks with balanced performance and reliability using a cloud browser. It handles complex interactions effectively with practical reasoning and clear execution. Optimal for tasks requiring: - Reliable, general-purpose browser automation - Clear, structured interactions with moderate complexity - Efficient handling of common web tasks and workflows Best suited use cases include: - Standard multi-step registration or form submissions - Navigating typical web applications requiring multiple interactions - Conducting structured web research tasks - Extracting data through interactive web processes Provide a clear step-by-step description, necessary context, and expected outcomes. Returns the completed result or an error message if issues arise.. It is categorised as a Destructive tool in the Hyperbrowser MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Hyperbrowser MCP server in PolicyLayer and add a rule for openai_computer_use_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 Hyperbrowser. Nothing to install.
openai_computer_use_agent is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the openai_computer_use_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 openai_computer_use_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.
openai_computer_use_agent is provided by the Hyperbrowser MCP server (hyperbrowserai/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 10 Hyperbrowser tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.