cdp_command

Call a single Chrome DevTools Protocol command on the current or specified tab.

SERVERAgent Browser SOURCE335234131/agent-browser-mcp

High RISK CLASS

Category Execute

Parameters 00 required

Recommended Rate-limitedsee the rule below

Registry record Grade F, identity unverified Pull the record →

This record as markdown: /tools/335234131-agent-browser-mcp/cdp-command.md

WHAT IT DOES

What cdp_command does on Agent Browser

AI agents invoke cdp_command to trigger actions in Agent Browser. 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.

RISK

Why cdp_command is rated High

CDP commands can execute arbitrary browser-level operations including JavaScript execution, network interception, DOM manipulation, file system access, and more. This tool enables full programmatic control of the browser at a low level, which goes beyond simple reads or writes. It could be used to exfiltrate cookies/sessions, inject scripts, navigate to malicious pages, or manipulate authenticated sessions.

From the tool's definition Call a single Chrome DevTools Protocol command on the current or specified tab

Attacks that exploit this kind of access

RECOMMENDED RULE

The rule that runs cdp_command safely

PolicyLayer is an MCP gateway: it sits between your AI agents and Agent Browser, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For cdp_command, this is the rule to start with:

cdp_command Rate-limited

cdp_command 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.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "cdp_command": {
      "limits": [
        {
          "counter": "cdp_command_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

RATE-LIMIT THIS TOOL → Instant setup, no code required.

The button opens the PolicyLayer dashboard: create your workspace, connect Agent Browser, apply this rule, and every cdp_command call is checked against it from then on.

FAQ

Questions about cdp_command

What does the cdp_command tool do? +

Call a single Chrome DevTools Protocol command on the current or specified tab. It is categorised as a Execute tool in the Agent Browser MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on cdp_command? +

Register the Agent Browser MCP server in PolicyLayer and add a rule for cdp_command: 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 Agent Browser. Nothing to install.

What risk level is cdp_command? +

cdp_command is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit cdp_command? +

Yes. Add a rate_limit block to the cdp_command 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.

How do I block cdp_command completely? +

Set action: deny in the PolicyLayer policy for cdp_command. 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.

What MCP server provides cdp_command? +

cdp_command is provided by the Agent Browser MCP server (335234131/agent-browser-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

KEEP EXPLORING

More on Agent Browser, and thousands of servers like it.

This server

All 20 Agent Browser tools → High-risk tools on this server →

Across the catalogue

adb_shell_command on Android Forensics ADB MCP Server → All high-risk MCP tools → The MCP Attack Database →

Guides

Govern code and CI agents →Agent sandbox →Least privilege for MCP →