AgentReadyHomeAgent Listing

← web-artifacts-builder

web-artifacts-builder — agentic threat model

8.4AIVSS 8.4 · High

This agent presents a high-risk profile due to its execution of local shell scripts and generation of complex frontend code, creating potential vectors for arbitrary code execution and supply chain injection if compromised.

OWASP AIVSS score rationale

AIVSS = (CVSS_Base + AARS) × Mitigation_Factor, where AARS = (10 − CVSS_Base) × (Factor_Sum / 10) × ThM
CVSS base 8.1AARS uplift 0.78Factor sum 3.9/10Threat ×1.05Mitigation ×0.95
Autonomy of Action
0.40
Goal-Driven Planning
0.50
Self-Modification
0.20
Dynamic Tool Use
0.70
Persistent Memory
0.10
Contextual Awareness
0.60
Dynamic Identity
0.10
Multi-Agent Interactions
0.20
Non-Determinism
0.60
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 — relies on Claude's underlying foundation model (Anthropic) which is susceptible to prompt injection that could manipulate the generated code or shell script execution parameters.

L2 · Data Operations⚠ not certain from listing

Not certain from the listing — the agent primarily processes code templates and user prompts rather than maintaining a persistent vector database or RAG pipeline.

L3 · Agent Frameworks✓ mapped

The agent framework orchestrates shell scripts (init-artifact.sh, bundle-artifact.sh) and invokes Parcel/Vite. Insecure tool integration here could allow an attacker to inject malicious shell commands or dependencies during the scaffolding and bundling phases.

L4 · Deployment & Infrastructure✓ mapped

The execution of shell scripts and local bundling via Parcel poses severe sandbox escape and host compromise risks if run in an unisolated local developer environment without strict containerization.

L5 · Evaluation & Observability⚠ not certain from listing

Not certain from the listing — there is no mention of built-in logging, guardrails, or AST-based security scanning of the generated React/TypeScript code before it is bundled and executed.

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

Not certain from the listing — as an open-source utility skill, it lacks native enterprise compliance controls, access policies, or automated vulnerability auditing for the 40+ preinstalled shadcn/ui components.

L7 · Agent Ecosystem✓ mapped

Designed as an official skill for Claude.ai artifacts; compromise of the hosting ecosystem or the artifact marketplace could lead to widespread distribution of malicious single-file HTML payloads.

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