make-com-israeli-automations
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill describes building automation workflows that ingest untrusted data from external sources such as Gmail attachments, WhatsApp messages, and various payment gateway webhooks. This ingested data is subsequently processed by AI Agents within the Make.com ecosystem, creating a surface for indirect prompt injection.
- Ingestion points: Data enters the workflow via 'Watch Emails', 'Watch Events', and 'Custom Webhook' modules (SKILL.md).
- Boundary markers: The instructions do not recommend using boundary markers or 'ignore embedded instructions' prompts to isolate external data from the AI agent's logic.
- Capability inventory: The workflows have the capability to create legal accounting documents in Morning and iCount, send outbound WhatsApp messages, and update CRM records in Monday.com.
- Sanitization: There is no mention of sanitizing or validating external inputs before they are processed by the AI Agent context.
Audit Metadata