AgentReadyHomeAgent Listing

← InboundCollie

InboundCollie — agentic threat model

8.7AIVSS 8.7 · High

InboundCollie presents a moderate-to-high risk profile due to its autonomous public-facing capabilities, specifically 24/7 voice telephony and automated landing page generation, which could be exploited for vishing, social engineering, or PII exfiltration if compromised.

OWASP AIVSS score rationale

AIVSS = (CVSS_Base + AARS) × Mitigation_Factor, where AARS = (10 − CVSS_Base) × (Factor_Sum / 10) × ThM
CVSS base 7.5AARS uplift 1.23Factor sum 4.7/10Threat ×1.05Mitigation ×1.0
Autonomy of Action
0.80
Goal-Driven Planning
0.50
Self-Modification
0.10
Dynamic Tool Use
0.60
Persistent Memory
0.70
Contextual Awareness
0.50
Dynamic Identity
0.20
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.

L1 · Foundation Models⚠ not certain from listing

Not certain from the listing — likely relies on third-party LLMs and TTS/STT APIs. Key threats include voice-based prompt injection (jailbreaking via phone) and model output misalignment during live customer interactions.

L2 · Data Operations⚠ not certain from listing

Not certain from the listing — stores customer PII, call transcripts, and qualification scores. Main threats include unauthorized database access, lack of encryption for voice recordings, and potential data leakage via the dashboard.

L3 · Agent Frameworks⚠ not certain from listing

Not certain from the listing — orchestrates telephony, landing page generation, and follow-up actions. Vulnerable to insecure tool integration where an attacker manipulates the agent into sending unauthorized SMS/emails or modifying landing page content.

L4 · Deployment & Infrastructure⚠ not certain from listing

Not certain from the listing — likely hosted on cloud infrastructure with telephony API integrations. Threats include insecure dashboard endpoints, exposed API keys for telephony services, and lack of isolation between tenant data.

L5 · Evaluation & Observability✓ mapped

The platform provides a dashboard with call transcripts and qualification scores for user review. However, it lacks visible automated guardrails to detect and block adversarial prompt injections or toxic outputs in real-time during live calls.

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

Not certain from the listing — handling voice recordings and PII introduces strict compliance requirements (GDPR, CCPA, TCPA). There is no mention of role-based access control (RBAC), data retention policies, or security certifications.

L7 · Agent Ecosystem⚠ not certain from listing

Not certain from the listing — operates primarily as a standalone vertical solution. Ecosystem risks are low unless it integrates with external CRMs, which could introduce cascading data synchronization vulnerabilities.

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