The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The file scripts/desform_jimureport.py contains a hardcoded cryptographic secret SIGNATURE_SECRET ('dd05f1c54d63749eda95f9fa6d49v442a') used to generate MD5 signatures for JimuReport API requests. This secret is an internal credential for the vendor's signature protocol.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted form design data and record content from external API endpoints, which are then interpolated into the agent context.
Ingestion points: API responses in scripts/desform_utils.py (via query_form, get_form_fields) and scripts/desform_data_utils.py (via list_data).
Boundary markers: Absent. The skill does not use delimiters or instructions to separate external data from its own core logic.
Capability inventory: The skill possesses network communication capabilities (urllib.request) and filesystem access (via skill_temp_path.py and export_design_json).
Sanitization: Absent. There is no filtering or validation logic for the content retrieved from the API before it is processed by the AI.
[DYNAMIC_CONTEXT_INJECTION]: The SKILL.md file uses the exclamation mark + backtick syntax (!python ...) to execute the scripts/gen_widget_options_index.py script at skill load time. This is used to dynamically build a widget options documentation index.
[COMMAND_EXECUTION]: Several Python scripts perform operations that interact with the system environment, such as creating temporary directories in scripts/skill_temp_path.py and generating MySQL commands for execution in scripts/desform_utils.py.
[OTHER]: The scripts scripts/desform_utils.py and scripts/desform_jimureport.py explicitly disable SSL certificate verification by setting ssl.CERT_NONE and check_hostname = False. This configuration makes the API communication vulnerable to man-in-the-middle (MitM) attacks.

jeecg-desform