oracle-database
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and provides instructions for downloading various development utilities and frameworks from well-known and trusted sources:
- Fetches the
utPLSQLtesting framework fromgithub.com/utPLSQL/utPLSQL.git. - Downloads the
utPLSQL-cliutility fromgithub.com/utPLSQL/utPLSQL-cli. - Retrieves
Swagger UIfor API documentation fromgithub.com/swagger-api/swagger-ui. - Downloads the official Oracle
SQLclutility fromdownload.oracle.com. - [PROMPT_INJECTION]: The skill includes defensive instructions and awareness regarding prompt injection:
- In
features/dbms-vector.md, the documentation explicitly warns about the risk of prompt injection in RAG (Retrieval Augmented Generation) pipelines and recommends input sanitization at the application layer. - In
ords/ords-sample-app.md, the skill includes meta-instructions that guide the AI agent to use only repository files and user-provided context as the source of truth, effectively acting as a safety boundary for the agent's behavior during complex configuration tasks.
Audit Metadata