Skip to content
#

data-freshness

Here are 9 public repositories matching this topic...

通用深度长文引擎(学术论文/商业评论/行业分析/公众号深度长文):八角色 T0-T8 + 三检索员并行 T1∥T2∥T6 + 三角验证 + M 机械硬门 + F 失败模式防御 + 批判伙伴 T8 + 修订回环≤2轮 + 数据信任3档 + 阶段闸门 T2.5/T5.5 + 内联引用支持 + 版本升级自审门 v2.2.8(7门机械化+按需加载模式+精简SKILL.md -37%/审计员卡 -77%/M门合并完整版,主流程读取 tokens -31%)。论衡哲学:独立角色+机械化门 > 自我复核

  • Updated Aug 19, 2026
  • Shell

Learns each dataset's own arrival distribution instead of applying one fixed freshness rule to everything: 86% fewer false alarms than an "alert if not loaded by 07:00" rule at 99.4% detection, plus lineage-aware alert suppression and per-dashboard trust scores. DuckDB, Python, Airflow-shaped, zero cloud credentials.

  • Updated Aug 3, 2026
  • Python

Declarative, scheduled refresh pipeline for map data: pull GeoJSON/JSON/CSV from HTTP APIs, files or SQL, transform and validate it, detect real changes with an order-stable content hash, and emit a freshness manifest plus a stale-data badge. Includes a composite GitHub Action that rebuilds only when the data moved.

  • Updated Jul 19, 2026
  • Python

Improve this page

Add a description, image, and links to the data-freshness topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the data-freshness topic, visit your repo's landing page and select "manage topics."

Learn more