skills/sales-skills/sales/sales-sonix/Gen Agent Trust Hub

sales-sonix

Pass

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is a legitimate informational tool for the Sonix platform. All referenced URLs are official and contextually appropriate.
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface through its persistent learning feature. 1. Ingestion points: The skill reads from references/learnings.md to load past platform knowledge. 2. Boundary markers: There are no explicit delimiters for the learned content. 3. Capability inventory: The agent can append new findings to references/learnings.md and route users to other sales skills. 4. Sanitization: The content added to the learnings file is not sanitized. This architecture is standard for knowledge-retrieval skills and is assessed as safe within the context of the skill's primary purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 22, 2026, 01:05 PM