Dovira AI — agentic threat model
Dovira AI is a low-autonomy, human-in-the-loop resume and interview preparation assistant. Its primary security risks center on the handling of sensitive personally identifiable information (PII) and the potential for indirect prompt injection via untrusted job descriptions.
OWASP AIVSS score rationale
| Autonomy of Action | 0.20 | |
| Goal-Driven Planning | 0.20 | |
| Self-Modification | 0.00 | |
| Dynamic Tool Use | 0.10 | |
| Persistent Memory | 0.30 | |
| Contextual Awareness | 0.40 | |
| Dynamic Identity | 0.00 | |
| Multi-Agent Interactions | 0.00 | |
| Non-Determinism | 0.50 | |
| Opacity & Reflexivity | 0.30 |
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 — likely utilizes commercial or open-source LLMs for resume rewriting and interview prep. Primary threats include indirect prompt injection where a malicious job description hijacks the model to output malicious payloads or exfiltrate user data.
Not certain from the listing — processes highly sensitive user PII (resumes, work history, contact details). Threats include data exfiltration of user profiles and potential data leakage if user inputs are used to train or fine-tune underlying models without consent.
Not certain from the listing — likely uses a simple chat-based orchestration framework. Threats include insecure handling of user-provided URLs if the agent attempts to scrape job descriptions directly from external job boards.
Not certain from the listing — hosted as a web application. Threats include standard web vulnerabilities, unauthorized access to user accounts, and lack of isolation between user sessions containing sensitive resume data.
Not certain from the listing — no mention of guardrails or observability tools. Threats include a lack of monitoring for abusive inputs or prompt injection attempts targeting the resume generation pipeline.
Not certain from the listing — handles sensitive candidate PII but does not specify compliance with GDPR, CCPA, or data retention policies. Threats include regulatory non-compliance and lack of audit trails for data access.
Not certain from the listing — operates as a standalone consumer-facing agent with no explicit multi-agent or ecosystem integrations described.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).