jeecg-system
Fail
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: HIGHCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The file
scripts/system_utils.pycontains a hardcoded cryptographic secret (_SIGN_SECRET = 'dd05f1c54d63749eda95f9fa6d49v442a'). This secret is used to generate MD5 signatures for API requests to the/sys/dict/getDictItems/endpoint. Hardcoding security secrets is a high-risk practice as it often assumes the target system uses default configurations.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external JeecgBoot API endpoints, creating a potential surface for indirect prompt injection.\n - Ingestion points: Data fetched from JeecgBoot APIs via
_requestinscripts/system_utils.pyand processed inscripts/system_creator.py.\n - Boundary markers: Absent; there are no clear delimiters or instructions to ignore embedded commands in the retrieved data.\n
- Capability inventory: The skill has the capability to perform network operations (POST and PUT) to create or modify system data such as roles, users, and dictionaries.\n
- Sanitization: No sanitization or validation of the API data is performed before it is processed within the agent context.\n- [COMMAND_EXECUTION]: The
SKILL.mdfile instructs the agent to execute local Python scripts via the shell using thepythoncommand, passing user-provided API base URLs and access tokens as command-line arguments.
Recommendations
- AI detected serious security threats
Audit Metadata