mongodb-crud-operations

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill defines capabilities for retrieving and processing MongoDB documents without providing examples for boundary markers or input sanitization. This establishes a surface for indirect prompt injection where malicious data in the database could influence agent logic.
  • Ingestion points: Data retrieval through 'findOne' and 'find' operations described in SKILL.md.
  • Boundary markers: None identified in the provided documentation or code snippets.
  • Capability inventory: Includes document insertion, querying, updating, and deletion.
  • Sanitization: No validation or sanitization logic is present in the provided examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 09:38 PM