跳到技能内容
词映 AI导航
发现工具选型指南Skills 技能AI 快讯AI图片AI视频AI文档AI办公AI助手AI搜索AI编程AI音频AI设计AI智能体模型与开发AI学习模型评测AI提示词内容检测AI金融AI 3D扫描与识别其他工具跨境电商 AI
词映 AI导航
工具库Skills 技能
我的收藏
词映 AI导航
发现工具选型指南Skills 技能AI 快讯AI图片AI视频AI文档AI办公AI助手AI搜索AI编程AI音频AI设计AI智能体模型与开发AI学习模型评测AI提示词内容检测AI金融AI 3D扫描与识别其他工具跨境电商 AI
Skills directory/huytieu/COG-second-brain/memory-hygiene

Knowledge Base Hygiene Skill for ChatGPT: Verify Drift, Confidence and Last Checked Dates

Distinguish volatile paths, URLs, and identifiers from durable preferences. Compare supplied evidence and record verification dates, drift, and potential archive candidates.

中文

Install in ChatGPT

Requires a ChatGPT workspace where you can create skills. This is a text adaptation of the pinned GitHub source.

  1. Review the inputs and outputs in the pinned .claude/skills/memory-hygiene/SKILL.md.
  2. In ChatGPT, open Plugins, Skills, Create, and Create with editor.
  3. Write the skill rules using the workflow described here. Provide recorded claims, dates, current evidence, and the authorized scope.
  4. Save the skill and check the fictional example in a new chat.

Example task

A fictional handbook points to /guide/v1, while the current index confirms /guide/v2. An editorial preference still applies, and another system ID has no evidence. Draft an audit table.

What to check

Record the path change with its source, retain the preference, and mark the unresolved ID unverified without assigning a new value.

Original SKILL.md on GitHub · Updated 2026-11-13

Frequently asked question

Should an unreachable link immediately cause deletion?

Record the check and time, then use authorized alternative evidence if available. Keep it pending until there is a basis to archive or replace it.

Browse ChatGPT skills in the Chinese directory
词映 AI 导航Skills 技能
工具分类全部工具已本地化技能选型指南AI 快讯