constructive-data-modeling

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill utilizes official SDKs from the verified author 'constructive-io', including @constructive-io/node-type-registry and @constructive-db/sdk. These resources are consistent with the vendor's infrastructure.
  • [SAFE]: Database provisioning and authentication examples use standard patterns, such as Bearer tokens and local development endpoints (localhost). No hardcoded credentials or sensitive data exposure was detected.
  • [SAFE]: The skill manages database permissions (Row Level Security and SQL Grants) as part of its core functionality for data modeling, using standard PostgreSQL roles like 'authenticated'.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided metadata for schema definitions (table and field names). However, the underlying platform implements validation triggers (validate_field_type, validate_field_default) to ensure data integrity and mitigate exploitation via malformed input.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 01:06 PM
Security Audit — agent-trust-hub — constructive-data-modeling