multi-platform-apps-multi-platform
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses the
$ARGUMENTSvariable to pass user input into various sub-agent prompts, such as the backend architect and feature implementation tasks. - Ingestion points: The
$ARGUMENTSvariable is processed inSKILL.mdto define API contracts and feature specifications. - Boundary markers: The skill does not use specific delimiters to isolate user-provided feature descriptions within the sub-agent prompts.
- Capability inventory: The skill orchestrates other agents via the
Tasktool but does not directly invoke high-risk commands or network operations itself. - Sanitization: No explicit sanitization of the input variable is performed before interpolation.
- [NO_CODE]: The skill contains only documentation, workflow instructions, and prompts for sub-agents. It does not include any executable scripts, binaries, or configuration files that run code on the local host.
Audit Metadata