domain-scaffolder-frontend

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands defined in mode configuration files, specifically {type_check_command}, {build_command}, {lint_command}, and {test_command}. While these are intended for post-scaffolding validation, they represent a sink for arbitrary command execution if configuration files are modified by an attacker.
  • [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection through the processing of plan files.
  • Ingestion points: The agent reads shared.md, frontend.md, and flows.md from a variable {plan_root}/{slice}/ directory.
  • Boundary markers: Absent. The skill does not use delimiters or instructions to ignore potential commands embedded within the plan markdown files.
  • Capability inventory: The skill possesses the capability to write files to the local filesystem (scaffolding) and execute shell commands (validation scripts).
  • Sanitization: Absent. Data extracted from the plans is directly interpolated into generated TypeScript code, API services, and component logic without escaping or validation.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with external packages via the {auth_packages_root} configuration. It suggests using package managers like npm or pip to link local auth service packages, which introduces external code into the execution environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 02:39 PM