The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains standard instructional guidelines for its role as a Release Manager. There are no patterns suggesting attempts to bypass safety filters or override system constraints.
[DATA_EXFILTRATION]: No access to sensitive files, environment variables, or credentials was detected. The skill does not perform network operations to exfiltrate data.
[REMOTE_CODE_EXECUTION]: There are no patterns for downloading or executing scripts from remote sources. The skill body and instructions are static text templates.
[COMMAND_EXECUTION]: The skill does not contain any shell commands or subprocess calls. It explicitly defines its capability boundaries, stating it cannot execute deployment commands or server operations.
[EXTERNAL_DOWNLOADS]: The skill references an external repository (github:mindverse/skillhub) as a recommendation for further tools. This is a passive informational reference and does not involve automated downloads or execution.

product-launch