← Back to Skills Marketplace
19280
Downloads
3
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install openclaw-agent-browser
Description
Headless browser automation CLI for AI agents. Use when interacting with websites — navigating pages, filling forms, clicking buttons, taking screenshots, ex...
Usage Guidance
Install only if you are comfortable giving an agent broad browser-control capability. Avoid saving credentials or auth state unless necessary, protect any saved state files like passwords, do not attach it to your everyday Chrome profile, and review the global installer or use a pinned, isolated installation.
Capability Analysis
Type: OpenClaw Skill
Name: openclaw-agent-browser
Version: 1.0.0
The OpenClaw AgentSkills bundle is classified as suspicious due to the extensive and high-risk capabilities exposed by the `agent-browser` tool, as detailed in `SKILL.md` and `references/commands.md`. Specifically, the `agent-browser eval "..."` command allows the AI agent to execute arbitrary JavaScript within the browser context, enabling potential data exfiltration (e.g., cookies, local storage, sensitive DOM content) and unauthorized network requests. Furthermore, commands like `agent-browser upload`, `agent-browser download`, and `agent-browser --allow-file-access open file:///...` grant the agent broad access to the local file system, posing risks for data exfiltration or unauthorized file manipulation. While these are designed features for browser automation, they create severe vulnerabilities for prompt injection attacks against the AI agent, allowing it to perform actions far beyond its intended scope if not strictly controlled by the OpenClaw runtime.
Capability Assessment
Credentials
The command reference documents uploads, downloads, recording, arbitrary JavaScript eval, existing-Chrome attachment, and local file access; these are disclosed but under-scoped for an agent-facing skill.
Install Mechanism
The setup script installs agent-browser globally from npm without a pinned version, then runs agent-browser install to retrieve Chromium and possibly OS dependencies, with no confirmation or integrity controls.
Instruction Scope
The trigger is very broad and includes login, downloads, form submission, monitoring, and any programmatic web interaction without requiring explicit confirmation for sensitive actions.
Persistence & Privilege
The skill documents auth vault credential reuse and browser state save/load for cookies and localStorage, including auto-save/restore, without strong warnings or consent boundaries.
Purpose & Capability
The stated purpose coherently covers navigation, form filling, scraping, screenshots, downloads, and web testing; the more powerful commands are browser-automation related but materially expand user impact.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install openclaw-agent-browser - After installation, invoke the skill by name or use
/openclaw-agent-browser - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release — headless browser automation CLI for AI agents
Metadata
Frequently Asked Questions
What is Agent Browser?
Headless browser automation CLI for AI agents. Use when interacting with websites — navigating pages, filling forms, clicking buttons, taking screenshots, ex... It is an AI Agent Skill for Claude Code / OpenClaw, with 19280 downloads so far.
How do I install Agent Browser?
Run "/install openclaw-agent-browser" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Agent Browser free?
Yes, Agent Browser is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Agent Browser support?
Agent Browser is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Agent Browser?
It is built and maintained by bodietron (@bodietron); the current version is v1.0.0.
More Skills