awesome-design-md

Warn

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructions frequently use npx --yes getdesign@latest to download and run code directly from the NPM registry. This enables the execution of remote code at runtime without prior verification of the package content.
  • [EXTERNAL_DOWNLOADS]: The skill relies on external network requests to fetch the 'getdesign' package and various design templates from the VoltAgent ecosystem.
  • [COMMAND_EXECUTION]: The workflow involves running shell commands to list templates, install design files, and verify filesystem state, which can be exploited if the inputs (like design slugs) are manipulated.
  • [PROMPT_INJECTION]: The skill introduces a risk of indirect prompt injection by ingesting externally provided DESIGN.md files and instructing the agent to treat them as a 'source of truth' for implementation decisions.
  • Ingestion points: The DESIGN.md file is written to the project's root directory via an external tool.
  • Boundary markers: Absent. There are no delimiters or instructions to ignore potential commands embedded within the design file.
  • Capability inventory: The agent possesses file system access, shell execution capabilities, and network access.
  • Sanitization: The skill does not include any validation or sanitization steps for the content retrieved from the external design repository.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 29, 2026, 10:05 AM