domains / data / training-health-monitoring / training-check
The two copies are nearly identical. The wanshuiyin copy is the authentic source because it contains the full, self-consistent procedure including the cron-wiring note and the specific reviewer model (gpt-6-astra) with xhigh reasoning effort. The brycewang copy is a mirror that has been altered, changing the model to gpt-5.4 and the reasoning effort to high, and removing the explanatory note about the skill's cadence. The core workflow, judgment table, and integration rules are identical, making the wanshuiyin version the definitive procedure.
| Repository | Path |
|---|---|
| wanshuiyin/Auto-claude-code-research-in-sleep | skills/training-check/SKILL.md |
| Role | Repository | Path |
|---|---|---|
| inferior | brycewang-stanford/Auto-Empirical-Research-Skills | skills/42-wanshuiyin-ARIS/skills/training-check/SKILL.md |
| adapter | wanshuiyin/Auto-claude-code-research-in-sleep | skills/skills-codex/training-check/SKILL.md |
| adapter | brycewang-stanford/Auto-Empirical-Research-Skills | skills/42-wanshuiyin-ARIS/skills/skills-codex/training-check/SKILL.md |