skills/bastndev/skills/start-astro/Gen Agent Trust Hub

start-astro

Pass

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill automatically detects the current directory name to define a PROJECT_NAME variable, which is then interpolated into generated project files like src/consts.ts and README.md.
  • Ingestion points: The project name is derived from the current working directory's basename.
  • Boundary markers: Absent; the value is placed directly into template strings.
  • Capability inventory: Includes file system writes and project build execution via 'bun run build'.
  • Sanitization: No sanitization is performed on the folder name. While this represents a surface for indirect prompt injection if an adversary provides a malicious folder name, it is standard behavior for development scaffolding tools.
  • [COMMAND_EXECUTION]: The skill performs shell commands to initialize and verify the project. Evidence includes calls to 'bun create astro@latest', 'bun install', and 'bun run build' in SKILL.md. These are standard operations within the Astro and Bun ecosystems.
  • [EXTERNAL_DOWNLOADS]: The skill downloads the Astro CLI and the @lucide/astro package. These resources are fetched from the official npm registry and represent well-known, trusted developer dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 16, 2026, 07:34 PM
Security Audit — agent-trust-hub — start-astro