AgentReadyHomeAgent Listing

← GPT-trainer

GPT-trainer — agentic threat model

9.4AIVSS 9.4 · Critical

GPT-trainer is a highly exposed multi-channel AI agent platform with significant risk stemming from its support for arbitrary REST APIs and function calling across public communication channels (SMS, WhatsApp, Social Media) without explicit built-in guardrails.

OWASP AIVSS score rationale

AIVSS = (CVSS_Base + AARS) × Mitigation_Factor, where AARS = (10 − CVSS_Base) × (Factor_Sum / 10) × ThM
CVSS base 8.5AARS uplift 0.88Factor sum 5.6/10Threat ×1.05Mitigation ×1.0
Autonomy of Action
0.70
Goal-Driven Planning
0.50
Self-Modification
0.10
Dynamic Tool Use
0.80
Persistent Memory
0.60
Contextual Awareness
0.70
Dynamic Identity
0.40
Multi-Agent Interactions
0.30
Non-Determinism
0.80
Opacity & Reflexivity
0.70

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.

L1 · Foundation Models✓ mapped

Supports major foundation models (GPT, Claude, Gemini). The primary threat is model misalignment and prompt injection vulnerabilities that can bypass system instructions across different model providers.

L2 · Data Operations⚠ not certain from listing

Not certain from the listing — likely utilizes vector databases for RAG to enable personalized business context, introducing risks of data poisoning, knowledge-base exfiltration, and unauthorized access to sensitive business data.

L3 · Agent Frameworks✓ mapped

Supports function calling, tool use, and REST APIs for workflow automation. This introduces severe risks of tool misuse, unauthorized API execution, and downstream system compromise via prompt injection.

L4 · Deployment & Infrastructure⚠ not certain from listing

Not certain from the listing — deployment spans web, FB, Instagram, WhatsApp, and SMS. The underlying hosting infrastructure, secrets management for API keys, and sandboxing of function execution are not specified.

L5 · Evaluation & Observability⚠ not certain from listing

Not certain from the listing — there is no mention of built-in evaluation, monitoring, logging, or guardrail mechanisms to detect and block malicious inputs or anomalous tool executions.

L6 · Security & Compliance (cross-cutting)⚠ not certain from listing

Not certain from the listing — while white-labeling and multi-tenancy are supported, specific security compliance standards (e.g., SOC2, GDPR) and robust role-based access controls (RBAC) are not detailed.

L7 · Agent Ecosystem⚠ not certain from listing

Not certain from the listing — although users can deploy multiple agents for clients, explicit multi-agent orchestration, agent-to-agent trust boundaries, or marketplace risks are not defined.

MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).