design-review
Warn
Audited by Snyk on Nov 18, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The required workflow opens the user-specified target URL in Aside or the gstack browser and reads its rendered text, DOM, and console output, which is outsider-authored content the agent must actively fetch, so it is low risk.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (medium risk: 0.30). The decisive boundary is the optional design-detector engine binary, which the skill downloads from an individual-owned GitHub repository (pbakaus/impeccable) and then runs over project files after user consent. Its checksum is pinned in gstack, but the enforcing code is not in the supplied content and the publisher is not a recognized first-party organization, so the provenance is treated as reputable unofficial, giving medium.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (low risk: 0.30). The skill mostly edits project source and makes git commits inside the repo, which is ordinary work. The main machine-level concern is the optional impeccable engine download, which writes a third-party binary into the home directory, but it is gated behind an explicit user yes, checksum-pinned, and logged. Nothing here uses sudo, creates accounts, or touches system or SSH/systemd config, so I'd treat it as a low-to-moderate risk rather than a clear violation.
Issues (3)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata