The Agent Skills Directory

[Data Exposure & Exfiltration] (SAFE): The skill provides instructions for managing sensitive API keys and database IDs via environment variables stored in a .env file. Documentation explicitly warns users not to commit this file to version control and suggests rotating keys if compromised. (Evidence: references/setup-guide.md).
[EXTERNAL_DOWNLOADS] (SAFE): The integration targets the official Notion API (api.notion.com). It requires standard, reputable Python libraries (requests, pyyaml) which are generally acceptable for this use case. (Evidence: references/error-handling.md).
[COMMAND_EXECUTION] (SAFE): The skill includes documentation for curl commands and shell scripts. These are intended for developer-side configuration, testing, and manual troubleshooting rather than for automated, untrusted execution by the agent. (Evidence: references/api-reference.md).
[Dynamic Execution] (SAFE): The included Python script (scripts/rate_limiter.py) implements standard exponential backoff logic for API rate limiting. It contains no use of unsafe dynamic execution sinks such as eval(), exec(), or runtime compilation. (Evidence: scripts/rate_limiter.py).

notion-master