database-expert
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and review user-provided code, which represents a surface for indirect prompt injection findings.
- Ingestion points: The skill analyzes code snippets provided in the chat context for architectural review (SKILL.md).
- Boundary markers: No specific delimiters are defined to isolate external code from system instructions.
- Capability inventory: The skill has access to tools including Bash, Write, and Edit, which allow for file system modifications.
- Sanitization: There is no explicit sanitization or filtering logic applied to the input code being analyzed.
- [EXTERNAL_DOWNLOADS]: The documentation suggests the installation and execution of external packages via npx for database integration.
- The skill documentation points to @modelcontextprotocol/server-postgres and @modelcontextprotocol/server-sqlite (SKILL.md).
- These are recognized as well-known service packages for the Model Context Protocol ecosystem and are documented for standard configuration purposes.
Audit Metadata