sales-outreach-personalization

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted prospect data and external signals to generate personalization hooks.
  • Ingestion points: User-supplied prospect information, external signals verified via web browsing tools, and data pulled from integrated CRM or enrichment platforms (SKILL.md).
  • Boundary markers: No delimiters or instructions are used to isolate external data from agent instructions, increasing the likelihood that embedded instructions in data could be obeyed.
  • Capability inventory: The skill's capabilities are limited to text generation (outreach hooks) and referencing sibling skills for drafting; it does not possess direct code execution or file-writing capabilities.
  • Sanitization: No sanitization or validation of retrieved external content is performed before interpolation into output hooks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 09:18 PM
Security Audit — agent-trust-hub — sales-outreach-personalization