skills/nonlooped/roblox-suite/Release/Gen Agent Trust Hub

Release

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill automates the release workflow using standard development tools, including git, gh (GitHub CLI), and node. These are used for repository maintenance and publishing tasks which align with the skill's stated purpose.
  • [EXTERNAL_DOWNLOADS]: Project dependencies are installed using npm ci during the validation phase. This is a standard procedure for verifying project integrity before a release.
  • [COMMAND_EXECUTION]: Executes local repository scripts (scripts/generate-catalog-artifacts.mjs) and ecosystem-standard tools (stylua, selene, rojo) to validate code quality and build artifacts before tagging.
  • [PROMPT_INJECTION]: The generation of release notes from git log output creates a potential surface for indirect prompt injection; however, this risk is mitigated by the skill's requirement for user review and explicit confirmation before any data is published.
  • [SAFE]: The skill incorporates robust safety rules, including instructions to respect .gitignore, avoid committing secrets, and prohibiting the use of high-risk flags like --force on the main branch.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 06:21 PM
Security Audit — agent-trust-hub — Release