design-md

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches public DESIGN.md files via the getdesign CLI (see Workflow A and scripts/fetch_all_design_md.sh which run npx getdesign@latest add and populate ~/.claude/skills/design-md/library//DESIGN.md), and those fetched, untrusted third‑party DESIGN.md files (including their Section 9 "Agent Prompt Guide") are read by downstream /minoan-frontend-design and used to guide implementation, so external content can materially influence agent actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). This skill invokes remote npm CLIs at runtime (e.g., "npx getdesign@latest" and "npx @google/design.md"), which execute fetched code and download DESIGN.md files (that include an "Agent Prompt Guide" used directly by agents), and references the upstream repo https://github.com/google-labs-code/design.md — so external content is both executed and can directly control agent prompts.