MD2Word — agentic threat model
MD2Word is a client-side utility tool rather than an active AI agent, presenting minimal agentic risk. The primary security concerns are limited to client-side vulnerabilities such as Cross-Site Scripting (XSS) during markdown parsing or the generation of malicious document payloads.
OWASP AIVSS score rationale
| Autonomy of Action | 0.00 | |
| Goal-Driven Planning | 0.00 | |
| Self-Modification | 0.00 | |
| Dynamic Tool Use | 0.00 | |
| Persistent Memory | 0.00 | |
| Contextual Awareness | 0.00 | |
| Dynamic Identity | 0.00 | |
| Multi-Agent Interactions | 0.00 | |
| Non-Determinism | 0.00 | |
| Opacity & Reflexivity | 0.00 |
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 tool appears to use deterministic parsing libraries rather than a foundation LLM. If an LLM is used behind the scenes, it would be vulnerable to indirect prompt injection via malicious markdown input, but this is highly unlikely given the 'in-browser' description.
Not certain from the listing — There is no backend database, RAG, or vector store mentioned. Data operations are restricted to local browser memory, eliminating server-side data exfiltration risks but leaving local storage (if used) as a potential target.
Not certain from the listing — No agentic framework or orchestration code is described. The application functions as a single-purpose utility translator rather than an autonomous agent executing tools.
Not certain from the listing — The application is deployed as a web-based tool. Infrastructure threats are limited to client-side web vulnerabilities, CDN compromise, or malicious script injection on the hosting domain.
Not certain from the listing — No evaluation, logging, or observability guardrails are mentioned. The 'no uploads' privacy model suggests zero server-side logging, which prevents centralized monitoring of abuse.
Not certain from the listing — There are no identity, authentication, or access control mechanisms since no accounts are required. Compliance is simplified due to the lack of data collection, but no formal certifications are cited.
Not certain from the listing — The tool operates in isolation and does not participate in any multi-agent ecosystem or marketplace integrations.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology. Are you the vendor? Factual corrections are free.