Selenium Automation Skill

Automate browser tasks using Selenium, including form filling, web scraping, UI testing, button clicks, alert handling, and capturing screenshots.

This skill appears to do what it claims (Selenium automation) but comes with important operational risks: it will visit arbitrary URLs and can fill/submit forms and interact with pages (including authenticated portals), and it downloads browser drivers at runtime. Before installing, consider: (1) Do not supply real account credentials unless you trust the code — prefer temporary/test accounts. (2) Run the skill in an isolated environment (VM/container) with limited network access if possible. (3) Review the included Python scripts locally to confirm no unexpected network callbacks or hardcoded endpoints — the published files show no explicit exfiltration endpoints, but dynamic interactions can leak data to the target sites. (4) Be aware webdriver_manager will fetch binaries from the Internet; if your environment restricts external downloads, adjust accordingly. (5) If you need to use this skill in production, add explicit safeguards (domain whitelists, credential handling policies, and logs) and consider code hardening and input validation. If you want, I can list the specific CLI flags the scripts appear to accept and points in the code you should review carefully (e.g., where credentials are used or saved).

Type: OpenClaw Skill Name: selenium-automation-skill Version: 1.0.0 The skill bundle provides standard browser automation tools using Selenium for form filling, web scraping, and task time registration. The scripts (form_filler.py, time_logger.py, and web_scraper.py) use legitimate libraries like webdriver-manager and BeautifulSoup4 to perform their stated functions. There is no evidence of data exfiltration, malicious execution, or prompt injection; the code is well-structured and follows common automation patterns.