scaffold-cli

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION] (LOW): The skill runs shell commands including npm install, npx ultracite init, and npm run build as part of the scaffolding process. This is expected behavior for this use case.
  • [EXTERNAL_DOWNLOADS] (LOW): The skill triggers npm install, which downloads dependencies from the npm registry. The packages listed in the templates (e.g., commander, vitest, biome) are standard, well-known utilities in the Node.js ecosystem.
  • [PROMPT_INJECTION] (LOW): Categorized as Indirect Prompt Injection (Category 8). The skill interpolates unvalidated user input such as {{name}} and {{description}} into generated files intended for AI agent context (AGENTS.md, SKILL.md).
  • Ingestion points: Variable collection in SKILL.md Step 1.
  • Boundary markers: Absent in generated templates; user strings are inserted directly into markdown.
  • Capability inventory: The resulting project includes capabilities for subprocess execution and dependency management via package.json scripts.
  • Sanitization: No sanitization or escaping is performed on user-provided strings before they are written to the documentation files.
  • [DYNAMIC_EXECUTION] (LOW): The skill generates source code, compiles it using tsdown, and سپس executes the resulting binary (node dist/cli.js) to validate the scaffold. This is a primary function of the skill, and the execution is limited to the newly created local directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:44 PM