From hw
Summarizes the latest Hypo-Workflow evaluation report, including milestone outcomes, scores, warnings, and analysis ledger summaries. Supports /hw:report --view for specific reports.
How this skill is triggered — by the user, by Claude, or both
Slash command
/hw:reportThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
📌 输出语言规则:
📌 输出语言规则: 读取 config.yaml → output.language
Use this skill to summarize the latest generated report file.
.pipeline/reports/ 中存在报告,或当前状态指向最新报告.pipeline/state.yaml。output.language 和 output.timezone。/hw:report --view M<N>,则在 .pipeline/reports/ 或已存档的 Cycle 报告目录中定位匹配的报告,并加载完整的报告内容。不要将 reports.compact.md 用于 --view。--view 目标且 .pipeline/reports.compact.md 存在,则首先显示紧凑报告摘要列表。history.completed_prompts[-1].report_file.pipeline/reports/ 中的最新报告output.language 摘要:
prompt_state.analysis_summary.ledger_path 定位 ledger,并摘要:
references/completion-report-contract.md 的八个必需字段;如果源报告缺字段,明确标注为缺失,而不是静默省略。.pipeline/PROGRESS.md 存在,请在概念上保持其摘要一致,但不要从只读报告命令对其进行修改。/hw:report --view M3:加载并显示 Milestone M3 的完整报告。/hw:report:当可用时,列出 .pipeline/reports.compact.md 中的紧凑报告摘要;否则摘要最新报告。output.language;默认为 zh-CNoutput.timezone;默认为 Asia/ShanghaiHH:MM,跨天 DD日 HH:MMzh-CN / zh 映射到 templates/zh/report.md,将 en / en-US 映射到 templates/en/report.md,当本地化模板缺失时回退到 templates/report.mdreferences/completion-report-contract.md — required completion narrative fieldsreferences/evaluation-spec.md — score interpretationreferences/commands-spec.md — report selection behaviorreferences/progress-spec.md — progress summary relationshipreferences/analysis-ledger-spec.md — Analysis ledger source of truth and summary boundarySKILL.md — full reporting contextnpx claudepluginhub hypoxanthineovo/hypo-workflow --plugin hwShow current Hypo-Workflow progress when the user wants a concise status summary without mutating pipeline state.
Generates or regenerates a comprehensive deep-work session reports with session lifecycle, slice summary, evidence trail, TDD compliance, sensor pass-rate, model usage, and evaluation outcomes. Invoked via `/deep-report` or `/deep-status --report`.
Compares harness evaluation history: shows score trends, per-tier deltas, diminishing returns detection, grade projections, bilingual reports, and ASCII charts. Useful after 2+ evaluations.