maravilla-db

Pass

Audited by Gen Agent Trust Hub on May 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation for the Maravilla DB platform service. It includes examples for CRUD operations, index management, and vector search.\n- [PROMPT_INJECTION]: The skill exposes an interface for reading data from a remote database (e.g., db.find). This represents a potential surface for indirect prompt injection if the database contains attacker-controlled content.\n
  • Ingestion points: Data returned from db.find, db.findOne, and db.findSimilar operations in SKILL.md enter the agent's context.\n
  • Boundary markers: No specific boundary markers or 'ignore' instructions are included in the provided code snippets to distinguish database content from instructions.\n
  • Capability inventory: The skill allows database reads/writes and authentication context retrieval.\n
  • Sanitization: No evidence of data sanitization or validation is present in the documentation snippets.
Audit Metadata
Risk Level
SAFE
Analyzed
May 27, 2026, 12:11 PM
Security Audit — agent-trust-hub — maravilla-db