The Agent Skills Directory

[DATA_EXPOSURE_AND_EXFILTRATION]: The skill communicates exclusively with api.wanfangdata.com.cn. This domain belongs to the official Wanfang Data platform, which is consistent with the skill's stated purpose. No unauthorized data transmission or access to sensitive local files was detected.
[CREDENTIALS_UNSAFE]: The Python scripts (wf_data_query.py, wf_data_get_doc.py, and wf_data_vector_search.py) retrieve API keys (WFDATA_APP_KEY and WFDATA_APP_CODE) from environment variables. This is a secure and standard practice for managing sensitive configuration data, avoiding the risks of hardcoded secrets.
[COMMAND_EXECUTION]: The skill uses standard network libraries (aiohttp, requests) to perform API calls. There are no patterns of arbitrary command execution, shell injection, or privilege escalation attempts.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests user-provided search queries and returns external literature data, it functions as a standard search interface. The processed data is restricted to the primary goal of literature discovery, and no malicious interpolation patterns were found.

wfdata-search