The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the way it processes external data. * Ingestion points: News titles and summaries are fetched from Eastmoney and Sina APIs in analyser.py. * Boundary markers: No delimiters or protective instructions are used when including this external content in reports. * Capability inventory: The skill utilizes network requests to retrieve data and writes persistent logs to the local file system. * Sanitization: The fetched news strings are not validated or escaped before being presented to the user.
[DATA_EXFILTRATION]: The skill initiates network connections to well-known financial domains to perform its core functions. * It accesses hq.sinajs.cn and eastmoney.com for real-time stock quotes and historical K-line data. * These operations are documented as safe as they target established financial data providers necessary for stock monitoring.
[COMMAND_EXECUTION]: The control.sh script facilitates process management for the monitoring service. * It executes shell commands to start and stop the Python-based daemon and manages background processes using nohup.

stock-monitor