Luke Agent Browser Clawdbot

PassAudited by ClawScan on May 10, 2026.

Overview

The skill is a straightforward browser-automation guide, but it can control websites and reuse login sessions if you choose to run its commands.

This appears safe to install as an instruction-only browser automation skill if you trust and verify the separate agent-browser CLI. Before using it, decide which sites and accounts the agent may control, avoid using high-privilege personal sessions for testing, and protect any saved auth state files.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If used on sensitive sites, the agent could submit forms, change settings, or trigger purchases/posts depending on the user's session and instructions.

Why it was flagged

The skill exposes direct browser interaction commands that can click, type, and submit content on websites. This is central to the stated purpose, but it can have real effects on logged-in accounts.

Skill content
agent-browser click @e2
agent-browser fill @e3 "text"
agent-browser type @e3 "text"
agent-browser press "Enter"
Recommendation

Use it only for clearly scoped tasks and require explicit user confirmation before actions that buy, post, delete, send, or change account data.

What this means

Saved auth files could let later browser automation act as the logged-in user, and mishandling those files could expose sessions.

Why it was flagged

The documented workflow can persist and reload cookies/storage, which may represent authenticated website sessions.

Skill content
agent-browser state save auth.json        # Save cookies/storage
agent-browser state load auth.json        # Load (skip login)
Recommendation

Store auth state files securely, avoid sharing them, delete them when no longer needed, and use separate low-privilege test accounts when possible.

What this means

The reviewed skill text is benign, but the actual behavior depends on the npm package and downloaded browser/dependencies installed on the user's machine.

Why it was flagged

The skill is instruction-only but asks the user to install and run an external CLI and browser dependencies that are not included in the reviewed artifacts.

Skill content
npm install -g agent-browser
agent-browser install                     # Download Chromium
agent-browser install --with-deps         # Linux: + system deps
Recommendation

Verify the npm package and GitHub source before installing globally, prefer pinned versions where possible, and avoid running install commands with elevated privileges unless necessary.