一键导入
predictive-power
Use when recording, evaluating, or interpreting Focusa predictions, risk forecasts, calibration stats, or predictive next-action guidance.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Use when recording, evaluating, or interpreting Focusa predictions, risk forecasts, calibration stats, or predictive next-action guidance.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when preserving Focusa cognitive state, resuming after compaction/model switch/context overflow, linking evidence to Workpoints, using Focus State, work-loop, lineage/tree, metacognition, state-hygiene, or diagnosing Focusa tool readiness.
Use when Focusa is offline, degraded, stale, writer-conflicted, non-canonical, or tools return pending/blocked results that need recovery.
Use when preserving Focusa cognitive state, resuming after compaction/model switch/context overflow, linking evidence to Workpoints, using Focus State, work-loop, lineage/tree, metacognition, state-hygiene, or diagnosing Focusa tool readiness.
Use when preserving or recovering Focusa Workpoint continuity: checkpoint/resume after compaction, link evidence, resolve active objects, or run drift-safe handoffs.
Use when operating Focusa through direct CLI or HTTP API: health checks, release proof, endpoint troubleshooting, parity checks, or daemon verification.
Use when updating Focusa public docs, skills, evidence files, README links, version snapshot language, or release-proof documentation.
| name | predictive-power |
| description | Use when recording, evaluating, or interpreting Focusa predictions, risk forecasts, calibration stats, or predictive next-action guidance. |
Use this skill when recording, evaluating, or interpreting Focusa predictions.
focusa_predict_stats or focusa predict stats to inspect calibration.focusa_predict_recordfocusa_predict_evaluatefocusa_predict_stats