jeecg-codegen
Fail
Audited by Gen Agent Trust Hub on May 9, 2026
Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The
SKILL.mdfile contains multiple instances of shell commands using hardcoded database credentials (-uroot -proot) for MySQL CLI operations. - [COMMAND_EXECUTION]: The skill instructions direct the agent to construct and run Bash commands (e.g.,
mysql,find,ls) using variable interpolation for parameters like{dbname},{后端根路径}, and{project_vue_root}. The absence of explicit sanitization for these user-provided inputs creates a vulnerability to command injection. - [COMMAND_EXECUTION]: The skill defines a workflow to automatically execute generated SQL files against a local database (127.0.0.1/localhost) using the
mysqlcommand-line tool. - [COMMAND_EXECUTION]: The
findcommand is used with project-relative paths to scan for existing Java and TypeScript files, allowing the agent to read and modify local source code.
Recommendations
- AI detected serious security threats
Audit Metadata