swiftui-design-consultation

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses eval to source environment variables from a local utility (gstack-slug). This is a standard pattern within the gstack ecosystem for managing project-specific state.
  • [COMMAND_EXECUTION]: The skill executes shell commands for file management (mkdir, cp, mv) and git operations (add, commit) to manage project artifacts and state markers.
  • [PROMPT_INJECTION]: The skill ingests information from project files to contextualize design proposals.
  • Ingestion points: README.md, CLAUDE.md, and existing Swift code files indexed via MCP (Phase 1.1, Phase 6.6).
  • Boundary markers: Delimiters are not explicitly used for raw ingestion, but inputs are passed into a structured YAML proposal for user review.
  • Capability inventory: Bash (git, mkdir, cp, mv), Write/Edit (artifacts), and Skill (chaining).
  • Sanitization: The skill enforces strict structured YAML schema validation and type-checking before any artifacts are generated from ingested data.
  • [SAFE]: The skill chains to other utilities using the built-in Skill tool for HTML preview generation and HIG conformance reviews. These dependencies are integrated components of the design consultation workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 07:17 AM
Security Audit — agent-trust-hub — swiftui-design-consultation