crawling-a-site

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides numerous examples of using the crawlberg CLI tool. This tool is designed for web crawling and captures reachable pages as Markdown or JSON. The skill includes standard flags for polite crawling such as --respect-robots-txt, --rate-limit, and --stay-on-domain. While it allows the execution of shell commands, these are restricted to the vendor's own crawling tool (crawlberg) and are used for its intended purpose of site analysis and documentation capture.
  • [REMOTE_CODE_EXECUTION]: No patterns of remote code execution or untrusted script downloads were found. The skill utilizes a local CLI tool and provides instructions on its flags.
  • [DATA_EXFILTRATION]: The skill is designed to fetch data from websites, but it does not contain patterns for accessing sensitive local files (like credentials or SSH keys) or sending such data to external servers.
  • [SAFE]: The skill follows security best practices by encouraging users to set page caps, respect robots.txt, and implement rate limiting when crawling third-party sites. The tool crawlberg is a vendor resource related to the skill author 'xberg-io'.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 10:22 PM
Security Audit — agent-trust-hub — crawling-a-site