scaffold-cli

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads Node.js packages from the official NPM registry via npm install and npx during the project bootstrapping phase to install required dependencies.
  • [COMMAND_EXECUTION]: The skill executes various shell commands to set up the project environment, including git init, ln -s, npm install, and npx ultracite init. It also verifies the generated CLI tool by executing the resulting binary with node dist/cli.js.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface (multi-step chain) in the generated project:
  • Ingestion points: User-provided inputs for {{name}} and {{description}} are gathered in Step 1 of SKILL.md.
  • Boundary markers: No delimiters or safety instructions are used when interpolating these variables into file templates.
  • Capability inventory: The skill writes new agent skill files and documentation and executes shell commands in references/post-scaffold.md.
  • Sanitization: There is no evidence of sanitization or validation of the input strings before they are written to the generated SKILL.md and README.md files.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 02:15 AM