new-miniapp

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill uses npx -y create-ait-app@latest to download and execute the vendor's official project initializer. It also performs npm install to set up dependencies for the new project.
  • [COMMAND_EXECUTION]: It executes several shell commands for directory management and file manipulation. It also runs a local shell script (inject-project-guide.sh) to apply design guidelines to the scaffolded project. Standard text processing tools like sed and awk are used to modify configuration files.
  • [DYNAMIC_EXECUTION]: The skill employs node -e to run inline JavaScript for verifying the contents of package.json and checking the presence of installed modules within the project's node_modules directory.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts an <app-name> parameter from the user. To mitigate command injection risks, it applies a slugification process (converting to lowercase, replacing non-alphanumeric characters with hyphens) and validates the result against specific length and naming constraints before using it in shell arguments.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 12:11 PM