brand-bluewaves
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (LOW): The
assets/manifest.jsonfile references an external icon source (https://unpkg.com/@phosphor-icons/core@2.1.1/assets/regular). This involves a network request to a non-whitelisted domain to fetch brand assets. - [COMMAND_EXECUTION] (LOW): The skill frontmatter requests
Bash,Read, andWritetool permissions. These are intended to facilitate document composition via the mentionedrender.pyandcompose.pyscripts. - [Indirect Prompt Injection] (LOW): The skill defines an attack surface where processed data could influence agent instructions. Evidence: (1) Ingestion points:
assets/manifest.json,assets/templates/pdf/zones.json. (2) Boundary markers: None identified. (3) Capability inventory:Bash,Read, andWritetools. (4) Sanitization: No sanitization or validation logic is present in the provided configuration files.
Audit Metadata