GHL AI Agents
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill includes benign prompt templates designed for business receptionist roles. These examples do not attempt to bypass safety guidelines or manipulate the analyzing agent's behavior.
- [PROMPT_INJECTION]: The documentation describes AI systems that ingest data from untrusted sources, representing an indirect prompt injection surface.
- Ingestion points: SMS, Facebook, Instagram, and Website URLs (SKILL.md).
- Boundary markers: Prompt templates include instructions to 'Never make up information' and adhere to specific business rules.
- Capability inventory: Appointment booking via synced calendars and call routing/transfer.
- Sanitization: The guide recommends testing with 'adversarial questions' to proactively find and fix behavioral weak spots.
- [NO_CODE]: The skill consists entirely of markdown documentation and does not contain any executable code, scripts, or remote download commands.
Audit Metadata