ruoyi-code-generator
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted user input to generate functional code and database scripts, which creates a potential injection vector if the input is malicious.
- Ingestion points: User-provided table metadata defined in
SKILL.md, specifically thetableCommentand thecommentfield for each entry in thecolumnsarray. - Boundary markers: The skill does not implement boundary markers or instructions to the LLM to treat the user-provided metadata as literal strings that should not be interpreted or parsed.
- Capability inventory: The skill generates a wide range of sensitive artifacts, including Java Controllers (
templates/java/controller.java.vm), Domain entities (templates/java/domain.java.vm), MyBatis XML mappings (templates/xml/mapper.xml.vm), and SQL menu initialization scripts (templates/sql/sql.vm). - Sanitization: There is no evidence of sanitization or escaping mechanisms for the user-provided comments. If a user provides a comment containing SQL injection payloads or Java code snippets, these could be interpolated directly into the generated files, potentially leading to security vulnerabilities in the application that uses the generated code.
Audit Metadata