The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the workflow instructions or configuration. The skill effectively defines a structured process for managing backend engineering tasks.
[COMMAND_EXECUTION]: The skill is configured to use the Bash tool for standard development activities, such as executing local commands or managing files, which is consistent with its stated purpose of orchestrating backend implementation workflows.
[PROMPT_INJECTION]: The skill accepts user input for feature names (<기능명>). While this is a potential surface for indirect prompt injection when the description is passed to downstream agents, it is a standard design pattern for orchestration skills and includes verification phases (security audit, code review) to mitigate risks.

backend-implementation