vowel-webcomponent
Pass
Audited by Gen Agent Trust Hub on Mar 23, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs users to install
@vowel.to/clientand@ricky0123/vad-webviabun add. These are standard dependencies for the widget's voice activity detection and client-side logic. - [COMMAND_EXECUTION]: Provides setup commands using
bunfor package management andcpfor copying assets to a public directory, which are standard development workflows. - [PROMPT_INJECTION]: The skill describes a surface for indirect prompt injection where the voice agent processes user speech to trigger actions via
registerAction. While this is an architectural risk inherent to voice-controlled agents, there are no malicious instructions in the skill itself. - Ingestion points: User voice transcripts processed by the
<vowel-voice-widget>(SKILL.md). - Boundary markers: None explicitly defined in the examples.
- Capability inventory: JavaScript action handlers registered via
registerActionorcustom-actionsattribute (SKILL.md). - Sanitization: Not explicitly mentioned in the integration examples.
Audit Metadata