browser-use

browser-use is a skill for Claude Code, Codex from QwenLM/qwen-code. It costs 21 tokens per session (1,473 once invoked), scanned A, original, Apache-2.0.

A browser-control integration that uses an existing Chrome session to observe and operate web pages. It relies on a Chrome extension and a local connection between the browser and the coding agent.

In plain words
What is it for?
Use it to build computer-use workflows that inspect Chrome and carry out bounded actions. A separate decision layer can choose from application-provided actions while the driver performs and verifies them.
Why use it?
It removes the need to start a separate browser-control service or install an unrelated extension. It also provides setup guidance when the required extension is missing.

Skill for Claude CodeCodex ✓ vendor

Written for no agent in particular: nothing here depends on one.

Good fit Use it to build computer-use workflows that inspect Chrome and carry out bounded actions. A separate decision layer can choose from application-provided actions while the driver performs and verifies them.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/qwenlm/qwen-code/browser-use
About the project

Qwen Code is an open-source AI coding agent that runs in a terminal and helps developers work with code through language models. It supports multiple model providers and can also be used through IDEs, desktop software, SDKs, and messaging bots.

QwenLM/qwen-code · 28,072 stars · on GitHub · qwenlm.github.io

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add QwenLM/qwen-code --skill browser-use
Clone the repo
git clone --depth 1 https://github.com/QwenLM/qwen-code

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for browser-use

README.md
[![agentmods](https://agentmods.dev/badge/skills/qwenlm/qwen-code/browser-use/github.svg)](https://agentmods.dev/skills/qwenlm/qwen-code/browser-use)
Your own site
<a href="https://agentmods.dev/skills/qwenlm/qwen-code/browser-use"><img src="https://agentmods.dev/badge/skills/qwenlm/qwen-code/browser-use/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for browser-use

Your own site · 80×15
<a href="https://agentmods.dev/skills/qwenlm/qwen-code/browser-use"><img src="https://agentmods.dev/badge/skills/qwenlm/qwen-code/browser-use.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 21 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,473 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00021 $0.01473
Opus 5.5 $0.00008 $0.00589
Sonnet 5 $0.00004 $0.00295
Haiku 4.5 $0.00002 $0.00147

Measured 2d ago against content hash 6696e1f6c8bc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-23, from the pricing page.

Security

Grade A, and why

browser-use scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (skill.test.ts), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

packages/core/src/skills/bundled/browser-use/SKILL.md · 151 lines

How it starts

The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Browser Use

Browser Use is bundled with Qwen Code and uses the generic Node REPL MCP tools and the Qwen Chrome extension. Do not install a separate Qwen extension or start a separate Browser Use MCP server.

Setup

Install and enable the Qwen Code Chrome extension in the profile the user wants to use. On macOS and Linux, first use of this SDK automatically registers the shared Native Messaging host in the user's installation directory. A browser task opts into this local setup, which can finish before the extension connects. The SDK checks the actual connection and protocol instead of reading Chrome's extension preferences.

If the extension does not connect, ask the user to open Chrome and check the extension in the intended profile. There is no store listing yet: build the extension from packages/chrome-extension in the Qwen Code repository (its README) and load dist/extension through chrome://extensions (Developer mode → Load unpacked). After installation or enabling it, retry the connection.

A usable Host of the same protocol is reused. A Host installed by a newer Qwen Code is never downgraded: if setup reports one, tell the user to update Qwen Code and stop. The Host files persist after Qwen exits. The user can inspect or remove them with node <skill-base>/runtime/scripts/native-host-setup.js status or uninstall; install explicitly switches the Host to this Qwen Code's copy from the next time Chrome starts it. A later Browser Use initialization can register the Host again. If required Host files cannot be read or written, report the failing action and path so the user can resolve access and retry.

Multiple Qwen sessions can use the same profile, each controlling its own tabs. A tab held by another session reports TAB_OWNERSHIP_CONFLICT. Use another tab or wait for that session to release it. The setup below binds the runtime to the default profile through browserAgent.browsers.get('chrome'). Only when the user names a specific Chrome profile, call browserAgent.browsers.list() before that line and pass the returned ID to browserAgent.browsers.get(id) instead. The runtime stays bound to the first profile it connects to; to switch profiles, call node_repl_reset and run the setup again.

Read the full file on GitHub · 151 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago Changed · +14 lines 6696e1f6c8bc
  2. 4d ago First seen · 137 lines · 21 tokens per session scan A 3b3c58285acb

Subscribe to this mod's changes

browser-use is a skill published in the GitHub repository QwenLM/qwen-code (28,072 stars, last pushed today), licensed Apache-2.0. It adds 21 tokens to every session and 1,473 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-19.