Security audit
Security checks for vulnerabilities and agentic risk
The skill is focused on creating DESIGN.md files, but it asks the agent to run unpinned npm tooling that can change after review.
Review before installing if your agent runs skills with broad filesystem or network access. Prefer a pinned, locally installed `@google/design.md` version or a sandboxed run, and confirm the agent only writes DESIGN.md.
The skill instructs use of npx @google/design.md spec without pinning an exact package version, which allows whatever version is current at execution time to be downloaded and run. In an agent context, that creates a supply-chain risk: a compromised upstream package, malicious new release, or dependency confusion-style event could lead to arbitrary code execution on the host running the skill.
The command npx @google/design.md lint DESIGN.md executes an unpinned npm package, meaning the exact code run is not fixed by the skill. Because npx may fetch and execute remote package code, an attacker who compromises the package or a future release could gain code execution during validation.
The skill runs npx @google/design.md export --format <format> DESIGN.md without a pinned version, so execution depends on the latest available package contents at runtime. That exposes the agent to arbitrary code execution or tampered output if the package or its dependency chain is malicious or unexpectedly changed.
The update workflow includes npx @google/design.md diff <previous-file> DESIGN.md with no version pin, again introducing execution of mutable third-party code. In this skill, repeated validation and diff steps increase exposure because the agent is explicitly directed to invoke the package multiple times during normal operation.
No suspicious patterns detected.
