orbitant-voice-drafting
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data from external communication channels to generate replies, which creates a surface for indirect prompt injection attacks.
- Ingestion points: The skill ingests external content from email and Slack messages when executing the
/triageand/crmcommands, and optionally extracts voice profile data from the conversation context in Cowork sessions. - Boundary markers: There are no explicit instructions or delimiters defined to separate external message content from the agent's internal drafting instructions.
- Capability inventory: The skill is limited to generating draft text and explicitly forbids autonomous sending, requiring the user to review and confirm with a 'send' command.
- Sanitization: The instructions do not specify any sanitization or filtering logic to prevent embedded instructions in external messages from being followed.
Audit Metadata