The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documents the use of standard curl commands to interact with the MOEX API (apim.moex.com). These commands use a bearer token for authentication, which is common practice for REST APIs. There is no evidence of arbitrary or dangerous command execution.
[EXTERNAL_DOWNLOADS]: The skill references and fetches data from apim.moex.com. As this is the official domain for the Moscow Exchange API, it is considered a well-known and trusted service for financial data.
[DATA_EXFILTRATION]: No patterns of sensitive data exfiltration were detected. The skill focuses on retrieving public or authenticated market data. Instructions specifically advise against leaking credentials by using environment variables like ${APIKEY}.
[PROMPT_INJECTION]: The instructions do not contain attempts to override agent safety guidelines. It includes a 'Boundary' section that explicitly restricts the agent from providing financial advice or predictions, which is a safety best practice.
[DYNAMIC_EXECUTION]: The skill provides code snippets for normalizing JSON data and implementing pagination. These are standard data processing logic patterns and do not involve eval(), exec(), or other unsafe dynamic execution methods.

algopack-mega-alerts