PlanetScale MCP Server — agentic threat model
The PlanetScale MCP Server exposes highly sensitive database operations, schema modifications, and SQL execution capabilities directly to LLM agents, presenting a high-severity risk posture if paired with autonomous execution without strict human-in-the-loop controls.
OWASP AIVSS score rationale
| Autonomy of Action | 0.60 | |
| Goal-Driven Planning | 0.50 | |
| Self-Modification | 0.10 | |
| Dynamic Tool Use | 0.90 | |
| Persistent Memory | 0.20 | |
| Contextual Awareness | 0.40 | |
| Dynamic Identity | 0.30 | |
| Multi-Agent Interactions | 0.30 | |
| 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.
Not certain from the listing — The MCP server itself is model-agnostic, but the underlying foundation model's susceptibility to prompt injection or jailbreaking directly translates to unauthorized SQL execution or schema alteration risks.
The agent operates directly on live database schemas, branches, and query insights. The primary threat is data exfiltration or unauthorized data modification via SQL execution tools exposed to the model.
The MCP framework orchestrates tool calling for database inspection and SQL execution. Insecure tool integration or lack of strict input sanitization on the generated SQL queries represents a critical vulnerability.
The server relies on a connected token's scope. Compromise of this token or the hosting environment of the MCP server allows direct, unauthenticated database access and potential lateral movement within the PlanetScale infrastructure.
Not certain from the listing — There is no mention of built-in guardrails, query validation, or anomaly detection to block destructive SQL commands (e.g., DROP TABLE) before they reach the database.
Access control is bound to the connected token's scope. If the token is over-privileged, the agent inherits full administrative capabilities, violating the principle of least privilege.
In a multi-agent setup, a compromised or rogue upstream agent could abuse this MCP server to execute malicious queries or alter database branches without direct human oversight.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).