The Agent Skills Directory

[SAFE]: The skill contains standard operational instructions and best practices for bot behavior with no malicious patterns detected.
[COMMAND_EXECUTION]: Provides a bash script template for implementing file-based locking to prevent duplicate cron job execution. This is a documented functional pattern for deduplication and does not involve unauthorized command execution or privilege escalation.
[EXTERNAL_DOWNLOADS]: The skill encourages a security-positive practice by advising the agent to use sandbox or MCP tools for HTTP requests instead of raw command-line tools like curl.

bot-defaults