oneshot-messaging

Pass

Audited by Gen Agent Trust Hub on Nov 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill allows the agent to ingest natural language instructions through several parameters, which could potentially contain malicious content aimed at manipulating the autonomous behavior of the voice agent or SMS sender.
  • Ingestion points: Untrusted data can enter the agent's context through the objective and context fields in agent.voice, as well as the message field in agent.sms.
  • Boundary markers: There are no explicit boundary markers or instructions to ignore embedded commands provided in the skill's implementation examples.
  • Capability inventory: The skill possesses the capability to communicate externally via SMS and Voice calls, including reading an SMS inbox.
  • Sanitization: The instructions do not specify any sanitization or validation routines to ensure that the input data does not contain prompt injection payloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 1, 2026, 06:34 PM