The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/domain_query.py uses subprocess.run("chcp 65001", shell=True) to set the console output to UTF-8 on Windows machines. This is a static, hardcoded command intended to fix character encoding issues and does not represent a command injection risk.\n- [DATA_EXFILTRATION]: The skill transmits user-provided domains and API credentials (JKHZ_ID, JKHZ_KEY) to the external service apihz.cn. This is the core purpose of the skill. While the use of query parameters for credentials is a design choice of the API provider, the communication occurs over HTTPS.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection if external data (WHOIS or ICP filing records) contains malicious instructions.\n
Ingestion points: Data enters the agent's context through the responses from cn.apihz.cn in scripts/domain_query.py.\n
Boundary markers: None are present; the API output is printed directly to stdout for the agent to read.\n
Capability inventory: The skill can execute Python scripts and make HTTP requests to its designated API endpoints.\n
Sanitization: The script filters raw control characters (\r and \x00) to improve JSON parsing stability but does not sanitize the semantic content of the retrieved records.

domain-query