AgentReadyHomeAgent Listing

← Alchemy MCP Server

Alchemy MCP Server — agentic threat model

7.9AIVSS 7.9 · High

The Alchemy MCP Server acts as a critical data and simulation gateway for blockchain-enabled agents, presenting high indirect risk; while read-oriented, its untrusted multi-chain outputs and transaction simulations can be manipulated to trick downstream agents into executing catastrophic financial transactions.

OWASP AIVSS score rationale

AIVSS = (CVSS_Base + AARS) × Mitigation_Factor, where AARS = (10 − CVSS_Base) × (Factor_Sum / 10) × ThM
CVSS base 7.5AARS uplift 0.87Factor sum 3.3/10Threat ×1.05Mitigation ×0.95
Autonomy of Action
0.30
Goal-Driven Planning
0.10
Self-Modification
0.00
Dynamic Tool Use
0.60
Persistent Memory
0.10
Contextual Awareness
0.50
Dynamic Identity
0.20
Multi-Agent Interactions
0.80
Non-Determinism
0.40
Opacity & Reflexivity
0.30

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 — The listing describes an MCP server providing tools rather than a specific foundation model. However, downstream models consuming this data are vulnerable to indirect prompt injection via malicious on-chain contract metadata or transaction traces.

L2 · Data Operations✓ mapped

Retrieves multi-chain blockchain data (tokens, NFTs, transfers). The primary threat is data poisoning or provenance gaps, as returned addresses and contract data are explicitly untrusted inputs that can be spoofed on-chain to mislead the consuming agent.

L3 · Agent Frameworks✓ mapped

Exposes transaction simulation and trace tools. Insecure tool integration is a major threat if downstream agent frameworks blindly trust simulated transaction success metrics to authorize real-world financial operations.

L4 · Deployment & Infrastructure✓ mapped

Operates as a remote MCP server requiring API keys. Threats include API key exposure, lack of transport layer security, and potential compromise of the remote hosting infrastructure serving the API.

L5 · Evaluation & Observability⚠ not certain from listing

Not certain from the listing — There is no mention of logging, monitoring, or guardrails to detect anomalous simulation requests, API abuse, or malicious payloads returned from the blockchain.

L6 · Security & Compliance (cross-cutting)✓ mapped

Uses API keys for metered access control, but lacks fine-grained authorization or input sanitization, passing untrusted blockchain data directly to the agent ecosystem without validation.

L7 · Agent Ecosystem✓ mapped

Designed specifically for agent-to-agent (A2A) and agent-to-tool ecosystems. A compromised or manipulated simulation output can cause cascading failures across multiple coordinated agents relying on the same blockchain state.

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