unity-vivox
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The documentation demonstrates a strong security posture by correctly identifying high-risk practices, such as shipping signing keys or secrets in client-side code, and providing explicit instructions to use server-side token minting instead.
- [EXTERNAL_DOWNLOADS]: The skill references official Unity packages (e.g.,
com.unity.services.vivox) and official documentation domains (docs.unity.com). These are well-known and trusted technology provider resources. - [INDIRECT_PROMPT_INJECTION]: The skill defines patterns for handling untrusted user chat data via
VivoxMessageobjects. - Ingestion points: Incoming messages are received via
ChannelMessageReceivedandDirectedMessageReceivedevents documented inreference/events-and-participants.md. - Boundary markers: While examples do not show explicit delimiters for the message body, the documentation distinguishes between display names and stable player identities.
- Capability inventory: The referenced SDK allows for network-based message transmission, player muting, and channel management.
- Sanitization: The skill correctly identifies that validation and content filtering (e.g., profanity checks) must be performed on the server side to be effective.
Audit Metadata