ruyi-github-safe-publish

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a well-documented and transparent security tool. It provides a complete reference of its 135 scanning rules and contains an extensive test suite (235 tests) to ensure reliable performance without malicious intent.
  • [DATA_EXFILTRATION]: While the skill reads project files and pushes code to GitHub, its primary purpose is to sanitize data to prevent exposure. It implements safety measures such as masking found secrets (e.g., ghp_***) in its output and reports, and automatically adding its own scan reports to .gitignore to prevent accidental commits of sensitive summaries.
  • [REMOTE_CODE_EXECUTION]: The skill uses a sub-agent for AI semantic scanning. The sub-agent is explicitly configured with a restricted, read-only capability set (limiting it to Read, Grep, Glob, and read-only Bash operations) to mitigate the risk of unauthorized code execution or modification during the analysis phase.
  • [COMMAND_EXECUTION]: The skill interacts with standard development tools such as Git and the GitHub CLI (gh). These commands are used for legitimate project management tasks like creating repositories and managing branches, all of which are subject to user confirmation in the interactive workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 12:27 PM