trove-lint

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

trove-lint

trove-lint

A memory that isn't inspected quietly rots.
未经检查的记忆会悄然失效。

When to use

使用场景

  • Periodic health check (after a stretch of captures/ingests, or weekly).
  • After an import, to see what arrived unlinked.
  • When recall quality feels off.
  • 定期健康检查(在一段时间的捕获/导入后,或每周进行一次)。
  • 导入完成后,查看哪些内容未建立关联。
  • 当召回质量下降时。

Process

流程

Step 1 — run

步骤1 — 运行

lint
(or read the
trove://lint
resource). Findings:
orphan_node
,
missing_evidence
,
duplicate_title
,
dangling_edge
.
执行
lint
命令(或读取
trove://lint
资源)。检测结果包括:
orphan_node
missing_evidence
duplicate_title
dangling_edge

Step 2 — triage

步骤2 — 分类处理

  • Orphans: real pages nothing links to. For each (or the top N),
    recall
    its title to find candidate hubs; propose
    connect { predicate: "mentions" | "part_of" }
    connections.
  • Missing evidence: agent-captured claims without citations. Propose evidence backfill (
    annotate
    against a source span) or an explicit inference note via
    remember
    .
  • Duplicate titles: read both; propose merging (repoint edges to the survivor, then the loser is soft-deleted server-side) — surface, don't auto-merge.
  • Dangling edges: endpoints deleted; propose
    forget
    .
  • 孤立节点:指没有任何链接指向的真实页面。针对每个(或排名前N的)孤立节点,调用
    recall
    获取其标题以寻找候选中心节点;建议使用
    connect { predicate: "mentions" | "part_of" }
    建立关联。
  • 缺失证据:Agent捕获的声明没有引用来源。建议进行证据回填(针对来源片段执行
    annotate
    ),或通过
    remember
    添加明确的推断说明。
  • 重复标题:查看两个重复内容;建议合并(将所有边指向保留的节点,然后在服务器端软删除被淘汰的节点)——仅展示建议,不自动合并。
  • 悬空边:指向已删除的端点;建议执行
    forget
    操作。

Step 3 — propose, then apply

步骤3 — 提出建议并应用

Present findings grouped with proposed fixes as a checklist; apply what the user picks. Bulk linking passes are fine once approved.
将检测结果与对应的修复建议整理为清单呈现;应用用户选择的修复方案。获得批准后,可批量执行关联操作。

Step 4 — confirm

步骤4 — 确认

Summary counts before/after; what remains open.
展示修复前后的统计数据;列出仍未解决的问题。

Notes

注意事项

  • Smoke-test debris (actors ending
    -smoke
    ) is cleaned with
    npm run db:clean:smoke -- --apply
    in the repo, not by hand.
  • Jobs pending/failed counts come with the lint report on the dashboard's health card.
  • 烟雾测试产生的冗余数据(名称以
    -smoke
    结尾的角色)需在仓库中执行
    npm run db:clean:smoke -- --apply
    清理,无需手动处理。
  • 待处理/失败任务的数量会随检查报告一起显示在仪表盘的健康卡片上。

Anti-patterns

反模式

  • Don't auto-fix without proposing — especially merges.
  • Don't link orphans to hubs on title similarity alone; read the node first.
  • Don't treat every warning as urgent; a leaf note with no links can be legitimate.
  • 请勿未经提议就自动修复——尤其是合并操作。
  • 请勿仅根据标题相似度就将孤立节点关联到中心节点;请先查看节点内容。
  • 请勿将所有警告都视为紧急问题;没有链接的叶子笔记可能是合理的。