The Agent Skills Directory

[SAFE]: The skill serves as a sequence orchestrator, delegating task logic to specific sub-skills (gitnexus-plan, gitnexus-work, gitnexus-review) located within the project repository.
[SAFE]: Implements a mandatory blocking gate (AskUserQuestion) between the planning and execution phases, ensuring human-in-the-loop control before expensive or impactful operations are performed.
[SAFE]: The skill lacks network capabilities, file system writes outside of task-specific planning documents (docs/plans/), and administrative command execution.
[SAFE]: The skill includes explicit instructions for the agent to avoid self-modifying its own governance thresholds based on live task performance, which is a safety best practice.
[SAFE]: No obfuscation, prompt injection attempts, or persistence mechanisms were detected.

gitnexus-lfg