AI agents invoke browser_go_back to trigger actions in Byob. 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 triggers a browser navigation action (history traversal), which is an external browser operation. While relatively benign, it executes a real browser action that changes the state of the browser tab, fitting the Execute category. The blast radius is low as it merely navigates back in history without creating, modifying, or deleting data.
From the tool's definition Go back one step in the browser history of the given tab and wait for the new page
Attacks that exploit this kind of access
Go back one step in the browser history of the given tab and wait for the new page. It is categorised as a Execute tool in the Byob MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Byob MCP server in PolicyLayer and add a rule for browser_go_back: 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 Byob. Nothing to install.
browser_go_back 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 browser_go_back 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 browser_go_back. 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.
browser_go_back is provided by the Byob MCP server (wxtsky/byob). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.