creating-projects

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes various shell commands to initialize projects and configure environments. This includes using moi init or mo init for repository creation, gh repo edit for GitHub settings, and npm/pnpm/cargo for dependency management.
  • [EXTERNAL_DOWNLOADS]: Fetches development tools and libraries from official package registries. It installs packages such as @liangmi/vp-config, vite-plus, and various Rust utilities (cargo-insta, dprint) from established sources. It also configures GitHub Actions using workflows from the voidzero-dev and oxc-project organizations.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it interpolates user-supplied strings (like project descriptions) and external data (like GitHub PR titles and bodies) into template files and CLI arguments. While this is a standard feature for a project generator, it is a surface where external instructions could be introduced into the project context.
  • [DYNAMIC_EXECUTION]: The skill generates project infrastructure files (e.g., package.json, Cargo.toml, CI workflows) and then executes validation scripts (just ready, vp run check) against the newly created files to ensure the environment is correctly set up.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 03:07 PM