Taxy AI — agentic threat model
Taxy AI presents a high-risk profile due to its execution of browser-level actions (like GitHub workflows and calendar scheduling) using GPT-4, making it highly susceptible to indirect prompt injection from untrusted web page content.
OWASP AIVSS score rationale
| Autonomy of Action | 0.70 | |
| Goal-Driven Planning | 0.80 | |
| Self-Modification | 0.10 | |
| Dynamic Tool Use | 0.90 | |
| Persistent Memory | 0.20 | |
| Contextual Awareness | 0.80 | |
| Dynamic Identity | 0.80 | |
| Multi-Agent Interactions | 0.10 | |
| Non-Determinism | 0.70 | |
| 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.
Uses GPT-4. Highly vulnerable to indirect prompt injection where malicious instructions embedded in web pages hijack the agent's execution flow.
Not certain from the listing — no explicit RAG or vector database is mentioned, though it dynamically processes active browser DOM data as its primary context.
Translates LLM outputs into browser actions (clicks, keystrokes). Vulnerable to tool misuse if prompt injection forces the agent to perform unintended actions like deleting GitHub repos.
Deployed as a local browser extension. Risks include local storage exposure of API keys and potential extension sandbox escape if DOM manipulation is poorly isolated.
Not certain from the listing — being in a research preview phase, it likely lacks robust real-time guardrails, execution logging, or drift detection.
Operates with the security context of the logged-in user's browser session. Lacks enterprise-grade access controls, policy enforcement, or audit trails.
Currently operates as a single-agent browser automation tool with no multi-agent coordination or marketplace ecosystem described.
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.