design-code-architecture

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) because it reads session state from local project documentation files to resume the design journey.
  • Ingestion points: The skill reads docs/DESIGN-CODE-ARCHITECTURE-PLAN.md and other documentation artifacts at the start of a session (SKILL.md, Operating Rule 1).
  • Boundary markers: None; the agent is not instructed to use delimiters or ignore embedded instructions within these files.
  • Capability inventory: The skill has the ability to write to the filesystem and invoke other skills by name (SKILL.md, Operating Rules 4 and 6).
  • Sanitization: None; the skill does not specify any validation or filtering of the file contents.
  • [COMMAND_EXECUTION]: The skill provides instructions to suggest shell commands to the user, such as npx skills add wondelai/skills/<slug> --global, to install missing architectural sub-skills. This is used for legitimate capability expansion from the author's verified repository ('wondelai').
  • [EXTERNAL_DOWNLOADS]: The skill references the download and installation of additional architectural tools from the vendor's own GitHub-hosted repository when required components are missing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 11:55 AM
Security Audit — agent-trust-hub — design-code-architecture