brainstorm-experiments-existing
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data provided by the user, including project descriptions and external files.
- Ingestion points: The skill utilizes user input (interpolated via
$ARGUMENTS) and refers to user-uploaded files such as PRDs, assumption lists, and designs. - Boundary markers: The instructions do not define specific delimiters or instructions to ignore potential commands embedded within user data.
- Capability inventory: The skill instructions do not invoke dangerous capabilities like network requests, file writing, or system command execution.
- Sanitization: No explicit sanitization or validation logic is present for the ingested data.
- [EXTERNAL_DOWNLOADS]: The skill provides links to product management resources hosted on
productcompass.pm. These are static reference links intended for human reading and do not trigger automated code execution.
Audit Metadata