Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计通过

memory-integrity-checker内存完整性检查器

Agent Skill

memory-integrity-checker 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

291

周安装

12

GitHub Stars

58

下载量

95
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:memory-integrity-checker(内存完整性检查器)
来源仓库:https://github.com/archieindian/openclaw-superpowers
仓库路径:skills/memory-integrity-checker
安装命令:
npx skills add https://github.com/archieindian/openclaw-superpowers --skill memory-integrity-checker
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/archieindian/openclaw-superpowers --skill memory-integrity-checker

简介

memory-integrity-checker 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。

  • 它适用于研究检索类任务,可帮助 Agent 验证数据完整性或筛选合规候选结果。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装,需结合原始 README 核验具体用法和功能边界。
  • 安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写操作。
  • 使用时需注意该技能属于研究检索类别,主要功能聚焦于信息定位而非直接执行系统级操作。

SKILL.md

Memory Integrity Checker

What it does

As memory DAGs grow through compaction, they can develop structural problems: orphan nodes with no parent, circular reference loops, summaries that inflated instead of compressing, broken lineage chains, and stale nodes that should have been dissolved. These problems silently corrupt the agent's memory.

Memory Integrity Checker runs 8 structural checks on the DAG, generates a repair plan, and optionally auto-fixes safe issues.

Inspired by lossless-claw's DAG integrity checking system, which detects and repairs corrupted summaries.

When to invoke

  • Automatically Sundays at 3am (cron) — weekly structural audit
  • After a crash or unexpected shutdown — check for corruption
  • When the agent's memory seems inconsistent — diagnose structural issues
  • Before a major compaction or prune operation — ensure clean starting state

Integrity checks (8 total)

CheckWhat it detectsSeverity
ORPHAN_NODENode with no parent and not a rootHIGH
CIRCULAR_REFCircular parent-child loops in the DAGCRITICAL
TOKEN_INFLATIONSummary has more tokens than its combined childrenHIGH
BROKEN_LINEAGEEdge references a node ID that doesn't existCRITICAL
STALE_ACTIVEActive node older than 30 days with no childrenMEDIUM
EMPTY_NODENode with empty or whitespace-only contentHIGH
DUPLICATE_EDGESame parent-child edge appears multiple timesLOW
DEPTH_MISMATCHNode's depth doesn't match its position in the DAGMEDIUM

How to use

python3 integrity.py --check                  # Run all 8 integrity checks
python3 integrity.py --check --fix            # Auto-fix safe issues
python3 integrity.py --check --only ORPHAN_NODE  # Run a specific check
python3 integrity.py --repair-plan            # Generate repair plan without fixing
python3 integrity.py --status                 # Last check summary
python3 integrity.py --format json            # Machine-readable output

Procedure

Step 1 — Run integrity checks

python3 integrity.py --check

Runs all 8 checks and reports findings by severity.

Step 2 — Review repair plan

python3 integrity.py --repair-plan

For each finding, shows what the auto-fix would do:

  • ORPHAN_NODE → reattach to nearest active root or deactivate
  • DUPLICATE_EDGE → remove duplicates
  • EMPTY_NODE → deactivate
  • STALE_ACTIVE → deactivate

Step 3 — Apply safe fixes

python3 integrity.py --check --fix

Auto-fixes LOW and MEDIUM severity issues. HIGH and CRITICAL require manual review.

State

Check results, finding history, and repair actions stored in ~/.openclaw/skill-state/memory-integrity-checker/state.yaml.

Fields: last_check_at, findings, check_history, repairs_applied.

Notes

  • Reads from memory-dag-compactor's state file — does not maintain its own DAG
  • Auto-fix only applies to LOW and MEDIUM severity issues
  • CRITICAL issues (circular refs, broken lineage) require manual intervention
  • Circular reference detection uses DFS with visited-set tracking
  • Token inflation check compares parent tokens vs. sum of children tokens

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.88%
按下载量换算32

Claude

29.85%
按下载量换算28

Cursor

18.18%
按下载量换算17

Gemini CLI

8.48%
按下载量换算8

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills