mongodb-connection
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided application environment details, workload descriptions, and existing code snippets to provide optimized configuration recommendations.
- Ingestion points: User-provided answers to context questions and code snippets processed at runtime (SKILL.md).
- Boundary markers: The instructions explicitly mandate asking one context-gathering question at a time and using logical justifications for all recommendations to maintain focus.
- Capability inventory: The skill contains no file-system write operations, network exfiltration commands, or subprocess execution capabilities.
- Sanitization: None specified beyond standard agent conversational guardrails; however, the skill's purpose is purely informational and lacks execution tools to exploit.
- [EXTERNAL_DOWNLOADS]: The skill provides links to official MongoDB documentation and technical specifications hosted on the MongoDB website and their official GitHub repository. These references are used solely for user information and do not involve automated script execution or package installation.
Audit Metadata