The Agent Skills Directory

[COMMAND_EXECUTION]: Ticker symbols provided as command-line arguments are validated against a strict regular expression (^[A-Z][A-Z0-9.\-]{0,11}$) in scripts/dividends.py, which prevents the injection of malicious payloads into the LLM prompt and ensures safe command-line processing.\n- [DATA_EXFILTRATION]: Network activity is restricted to the endpoint api.aisa.one using the official OpenAI client library. This domain is an established resource for the skill's author (aisa-team), and communication is strictly performed over HTTPS to protect data integrity.\n- [PROMPT_INJECTION]: The skill utilizes a defensive system prompt that explicitly defines the agent's role as a read-only researcher. It includes specific negative constraints to prevent the model from soliciting brokerage credentials or executing financial trades.

stock-dividend