retro-industrial-dashboard-design

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data including user-provided references and project source code, creating a potential attack surface for instructions embedded in data.
  • Ingestion points: The "Run focused discovery" section in SKILL.md directs the agent to inspect user-provided references, design files, existing DESIGN.md artifacts, component styles, and project entry points.
  • Boundary markers: The instructions lack explicit requirements for using delimiters or boundary markers to isolate ingested content from the agent's internal command context.
  • Capability inventory: The skill is capable of writing to DESIGN.md files and generating implementation code, providing a mechanism for an injection to produce persistent malicious output.
  • Sanitization: No sanitization, validation, or filtering of the ingested external content is described in the provided files.
  • [EXTERNAL_DOWNLOADS]: The skill maintains compatibility by referencing an external design specification from a trusted source.
  • Reference: The references/design-md-contract.md file points to the official Google Labs Code specification on GitHub (https://github.com/google-labs-code/design.md/blob/main/docs/spec.md) to handle format drift.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 12:00 PM
Security Audit — agent-trust-hub — retro-industrial-dashboard-design