mongodb-search-and-ai

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill incorporates robust safety principles by requiring the agent to explain planned index configurations in plain language and obtain explicit user consent before utilizing the create-index tool. This design pattern ensures that administrative actions are intentional and reviewed by a human operator.
  • [PROMPT_INJECTION]: The skill processes untrusted data from database collections, creating a theoretical surface for indirect prompt injection. Ingestion points: Database schema definitions and query result sets accessed through the collection-schema and aggregate tools (SKILL.md). Boundary markers: The instructions do not define specific data delimiters to isolate untrusted content from the system prompt. Capability inventory: The skill utilizes tools for both reading data (aggregate, collection-indexes) and modifying the schema (create-index) (SKILL.md). Sanitization: The workflow requires version checks and schema validation before any complex search operations are performed. Given the administrative context and mandatory approval step, this is evaluated as a low-risk, safe implementation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 05:08 PM