a1-migration
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts an optional theme argument which is interpolated into a shell command for execution. This represents a potential injection surface if the input is not strictly validated.
- Ingestion points: The
argument-hint('spacious | compact | dense') and the interactive prompt for theme selection in Step 1. - Boundary markers: No explicit delimiters or validation logic are provided to wrap the
<THEME>placeholder within the shell command. - Capability inventory: The skill executes Angular schematics using
ng g, which allows for shell command execution in the agent's environment. - Sanitization: The instructions do not specify sanitization or escaping of the user-provided theme input before it is passed to the CLI command.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute Angular CLI commands (
ng g @allianz/ngx-brand-kit:apply-a1-theme). These commands are scoped to the 'allianz' vendor and are part of the intended migration functionality, but they involve subprocess execution that could be exploited if the inputs are manipulated.
Audit Metadata