The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. The skill's functionality is well-documented and consistent with its stated purpose of managing agent loops.
[COMMAND_EXECUTION]: The skill facilitates the execution of terminal commands and file edits through its worker roles (Hermes). These capabilities are standard for an IDE-based workflow tool and are scoped to the development environment.
[PROMPT_INJECTION]: The skill is designed to process untrusted external data (such as command output and file content), which presents a potential surface for indirect prompt injection. This risk is mitigated by explicit instructions to escalate and require user approval for high-risk actions like credential usage, destructive changes, or production deployment. Ingestion points include file inspections and terminal results, while output is structured using YAML contracts and Markdown logs to maintain clarity.

loops