Jotform — agentic threat model
Jotform AI Agents present a moderate risk profile, primarily driven by their public-facing deployment on websites and integration with communication channels like WhatsApp. The main threats involve prompt injection leading to customer misinformation or data exfiltration of ingested company knowledge.
OWASP AIVSS score rationale
| Autonomy of Action | 0.40 | |
| Goal-Driven Planning | 0.20 | |
| Self-Modification | 0.10 | |
| Dynamic Tool Use | 0.30 | |
| Persistent Memory | 0.20 | |
| Contextual Awareness | 0.40 | |
| Dynamic Identity | 0.10 | |
| Multi-Agent Interactions | 0.10 | |
| Non-Determinism | 0.50 | |
| Opacity & Reflexivity | 0.50 |
Scored with the canonical OWASP AIVSS formula (AIVSS calculator reference); agentic risk factors estimated from the agent’s described capabilities.
MAESTRO 7-layer threat model
Per-layer threats for this agent. Layers tagged “not certain from listing” are general, caveated commentary where the public description didn’t pin that layer.
Not certain from the listing — the underlying LLM is not specified, but it is susceptible to prompt injection, adversarial manipulation, and mis-aligned outputs that could mislead customers.
The agent is trained on company data to assist users. Threats include knowledge-base poisoning if untrusted company data is ingested, and data exfiltration of proprietary company info via prompt extraction.
Not certain from the listing — the orchestration framework is proprietary. Risks include insecure tool integration with WhatsApp/Phone APIs and potential memory/session poisoning.
Not certain from the listing — deployment is managed by Jotform as a SaaS widget. Risks include cross-site scripting (XSS) if the embedded widget is compromised, and insecure API endpoints.
The listing highlights 'Form analytics features' but lacks explicit security guardrails or drift monitoring. Gaps in observability could allow prompt injection attacks to go undetected.
Not certain from the listing — while Jotform generally complies with privacy standards, the specific compliance posture (e.g., GDPR, HIPAA for agent data) is not detailed here.
The agent integrates with WhatsApp, Phone, and over 6000 templates. Risks include template-based supply chain vulnerabilities and trust abuse across communication channels.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).