Skip to main content
Glama

browser_open

Connect to a Chrome or Edge browser via remote debugging to get a list of open tabs with IDs, urls, and titles. Optionally launch a debug-mode browser automatically if none is running.

Instructions

Connect to Chrome/Edge running with --remote-debugging-port and return open tab IDs — required before all other browser_* tools. Pass launch:{} (or with overrides) to auto-spawn a debug-mode browser when no CDP endpoint is live (idempotent: an already-running endpoint is preferred). Returns tabs[] with id, url, title, active — pass tabId to browser_* tools to target a specific tab. Caveats: CDP connection is per-process; if Chrome restarts, call browser_open again to get fresh tab IDs. A Chrome session started without --remote-debugging-port cannot be taken over — close it first or use a separate userDataDir. If the CDP endpoint is unreachable and launch is omitted, returns ok:false (typically code:'BrowserNotConnected' when the fetch surfaces ECONNREFUSED, otherwise code:'ToolError' with error 'Cannot reach Chrome/Edge CDP...'); re-call with launch:{} (idempotent) to auto-spawn or start Chrome manually with --remote-debugging-port=9222.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
portNoChrome/Edge CDP remote debugging port.
launchNoIf set, spawn a debug-mode browser when no CDP endpoint is live on the target port (idempotent: an already-running endpoint is preferred and the spawn step is skipped). Pass {} to use defaults (chrome, C:\tmp\cdp, no initial URL). Omit to perform pure connect.
includeNoOptional response-shape opt-in. `['envelope']` returns the self-documenting envelope (`_version` / `data` / `as_of` / `confidence`). `['raw']` forces raw shape (overrides DESKTOP_TOUCH_ENVELOPE=1 server default). Default behaviour is raw shape (compat with existing clients).
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Discloses idempotency of launch, CDP per-process, warning about killExisting losing unsaved input, error codes (BrowserNotConnected, ToolError), and restart requirement. No annotations to contradict.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Single paragraph is efficient and front-loaded, but length is justified by complexity. Slightly dense, but no wasted sentences.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Comprehensive coverage of connect/launch semantics, error handling, caveats, and integration with sibling tools, all without an output schema.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Despite 100% schema coverage, the description adds high-level context: launch idempotency, default browser resolution, killExisting warning, and include parameter purpose, surpassing mere restatement.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states 'Connect to Chrome/Edge ... return open tab IDs — required before all other browser_* tools' and covers auto-spawn, distinguishing it from sibling tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly states it's required before other browser tools, provides when-to-use (connect or auto-spawn) and when-not (unreachable endpoint, need to close existing session), and error handling guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Harusame64/desktop-touch-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server