The Agent Skills Directory

[SAFE]: The skill defines a structured development workflow using standard toolsets. All identified shell commands are limited to local repository management (e.g., git status, git branch, git diff) and project structure validation (find).
[COMMAND_EXECUTION]: Shell commands are used for branch safety checks and verifying that the project tree is clean. These operations are performed on local files within the superRA/ directory and do not pose a security risk in the context of a development agent. The destructive action of deleting branches (Option 4) is protected by a mandatory user confirmation gate.
[REMOTE_CODE_EXECUTION]: The skill coordinates with other internal skills (e.g., superRA:agent-orchestration, superRA:superplan). These are platform-native modular components used for subagent dispatching and do not involve downloading or executing code from untrusted external sources.
[PROMPT_INJECTION]: Instructions provide coordination logic for implementer and reviewer agents. There are no attempts to override safety protocols, extract system prompts, or bypass AI behavioral constraints.
[DATA_EXPOSURE]: The skill interacts with the local Git repository and specific project directories (superRA/). It does not access sensitive user directories (like .ssh or .aws) or environment variables containing secrets.

superimplement