module-entity

Warn

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run shell commands (npm run test) with arguments like workspace names, module paths, and entity names derived from user input and files like package.json. This allows for potential command injection if these inputs contain shell metacharacters.
  • [PROMPT_INJECTION]: The skill reads various existing files (e.g., modules/auth/src/user/model/user.entity.ts, packages/shared/src/validation/rules/) to guide its code generation. This introduces a surface for indirect prompt injection where malicious instructions in the source code could influence agent behavior.
  • Ingestion points: Multiple project files in modules/, packages/shared/, and modules/<modulo>/package.json.
  • Boundary markers: None specified to distinguish code content from instructions.
  • Capability inventory: Reading local files, writing source/test files, and executing shell commands via npm.
  • Sanitization: No input validation or escaping is defined for variables used in shell command construction.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 23, 2026, 04:47 PM
Security Audit — agent-trust-hub — module-entity