The Agent Skills Directory

[SAFE]: The skill defines a robust, multi-phase autonomous workflow for project creation. Project state and configuration are managed locally within the .omc and .claude directories.
[COMMAND_EXECUTION]: The execution and QA phases involve standard software engineering tasks such as building, testing, and linting. These are expected behaviors for an autonomous coding agent and are performed within the local project context.
[PROMPT_INJECTION]: The skill ingests user descriptions to generate code. While this represents an indirect prompt injection surface, the risk is mitigated by built-in multi-agent validation steps, including dedicated security and architecture reviewers.

autopilot