Agent Browser

Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.

Install only if you need browser automation and trust the local agent-browser executable. Avoid using it on sensitive logged-in accounts unless necessary, confirm before actions that change data or upload files, treat saved state files and recordings as secrets, keep them out of repositories, and delete them when finished.

Type: OpenClaw Skill Name: agent-browser-3 Version: 0.1.0 The OpenClaw AgentSkills bundle is classified as suspicious due to the broad and high-risk capabilities of the `agent-browser` tool, even though no explicit malicious intent is demonstrated in the provided files. Key indicators include the ability to load custom browser executables and extensions (`--executable-path`, `--extension`, `AGENT_BROWSER_EXECUTABLE_PATH`, `AGENT_BROWSER_EXTENSIONS`), execute arbitrary JavaScript within the browser context (`agent-browser eval`), access local files via the `file://` protocol, and save/load full browser session state (`agent-browser state save/load`), which can contain sensitive authentication tokens. While these features are plausibly needed for browser automation, they present significant attack surface for potential misuse if the agent or user were compromised, despite the documentation's inclusion of security best practices.