This record as markdown: /tools/alexgladkov-claude-in-mobile/browser-open.md
What browser_open does on Claude Mobile
AI agents invoke browser_open to trigger actions in Claude Mobile. 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.
Why browser_open is rated High
Opening a URL in a browser session triggers an external operation on the device. While it reads/navigates content, the act of opening arbitrary URLs can lead to unintended consequences such as phishing pages, drive-by downloads, or initiating OAuth flows. It is best classified as Execute due to the external operation triggered on the device, with medium severity given the potential for misuse.
From the tool's definition 'Open URL in browser session' — triggers an external operation (opening a URL) on a mobile device browser
Attacks that exploit this kind of access
The rule that runs browser_open safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Claude Mobile, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For browser_open, this is the rule to start with:
browser_open 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.
The button opens the PolicyLayer dashboard: create your workspace, connect Claude Mobile, apply this rule, and every browser_open call is checked against it from then on.
Questions about browser_open
Open URL in browser session. It is categorised as a Execute tool in the Claude Mobile MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Claude Mobile MCP server in PolicyLayer and add a rule for browser_open: 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 Claude Mobile. Nothing to install.
browser_open 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_open 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_open. 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_open is provided by the Claude Mobile MCP server (alexgladkov/claude-in-mobile). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Claude Mobile, and thousands of servers like it.
Across the catalogue