Show HN: SkillCompass – open-source quality evaluator for your AI skills (github.com)

SkillCompass has been launched as an open-source tool designed to enhance the quality evaluation of AI skills, particularly for Claude Code and OpenClaw users. This local-first quality evaluator operates on a six-dimension scoring system, providing actionable insights to users by identifying and addressing the weakest areas in their skills. Instead of the traditional "tweak and hope" approach, SkillCompass promotes a systematic diagnostic process: evaluate skill health, pinpoint deficiencies, implement targeted improvements, and verify outcomes. This iterative loop encourages continuous development and adaptation of AI skills based on real usage data. The significance of SkillCompass for the AI/ML community lies in its capacity to offer ongoing visibility into skill performance without compromising data privacy, as all data remains local. The tool employs passive tracking of skill usage and automatically highlights which skills require attention, thus ensuring a focus on continuous improvement. Key features include structured evaluations across security, functionality, and uniqueness dimensions, a capability for automatic updates, and the ability to scan entire directories for comprehensive assessments. Designed to foster a culture of agile development, SkillCompass empowers developers to not only fix current issues but also evolve their skills with confidence.