Token导航 LogoToken导航TokenDH.com
效率external-serviceclawhub未标认证来源可访问clear审计通过

notestorelab-case-reviewnotestorelab 案例回顾

Agent Skill

用于辅助测试设计、自动化测试、用例整理和回归验证。它适合让 Agent 编写单元测试、端到端测试、测试计划或根据失败日志定位问题。使用时需要确认项目测试框架、运行命令和夹具数据,避免为了通过测试而改坏真实逻辑;涉及浏览器或外部服务时,应区分本地模拟、测试环境和生产环境。

总安装

3,374

周安装

142

GitHub Stars

公开资料未说明

下载量

1,181
OpenClaw

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

MIT-0

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:notestorelab-case-review(notestorelab 案例回顾)
来源仓库:https://github.com/xiaojiou176/notestorelab-case-review
安装命令:
openclaw skills install notestorelab-case-review
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install notestorelab-case-review

简介

当用户要求“查看 Apple Notes 案例根”、“设置 NoteStore Lab MCP”、“运行 Notes-recovery-mcp”、“询问一个有界问题”时,应使用此技能。

SKILL.md

name
notestorelab-case-review
description
This skill should be used when the user asks to "review an Apple Notes case root", "set up NoteStore Lab MCP", "run notes-recovery-mcp", "ask one bounded question about a case", or "compare two NoteStore Lab case roots". It teaches install, first-success proof, and bounded case review without turning the repo into a hosted platform.
triggers

NoteStore Lab Case Review

Use this skill when an agent needs to install, wire, or operate the NoteStore Lab review flow on one explicit case root or the public-safe demo surface.

What this skill teaches

  • how to prove the workflow shape on demo artifacts first
  • how to wire the local stdio MCP surface without pretending there is a hosted

service

  • how to inspect one case root at a time using derived artifacts first
  • how to ask bounded, evidence-backed questions instead of free-form guessing

MCP capability surface

  • read-only review surfaces: list_case_roots, inspect_case_manifest,

select_case_evidence, inspect_case_artifact, and ask_case

  • bounded workflows: run_verify, run_report, build_timeline, and

public_safe_export

  • one explicit case root at a time, local stdio only, with no live Notes store

access

Product truth

  • Recovery is the main product.
  • AI and MCP are review layers, not the recovery engine.
  • Stay local, copy-first, and case-root-driven.
  • Prefer derived artifacts before raw copied evidence.
  • Do not treat the live Notes store as a target.
  • Do not describe NoteStore Lab as a hosted or multi-tenant platform.

First-success flow

  1. Install or launch the shipped surface using references/install-and-mcp.md.
  2. Run the public-safe proof path:

- notes-recovery demo - notes-recovery ai-review --demo - notes-recovery ask-case --demo --question "What should I inspect first?" - notes-recovery doctor

  1. Only after the demo path works, point the MCP server at one explicit

Notes_Forensics_<run_ts> case root.

  1. Keep all reasoning on copied evidence and derived artifacts, not on the live

Notes store.

Preferred evidence order

  1. Read review_index.md.
  2. Read the verification preview and pipeline summary.
  3. Use notes-recovery ask-case for one bounded operator question.
  4. Use notes-recovery case-diff only when comparing two case roots.
  5. Use notes-recovery public-safe-export when you need a shareable bundle.

Bounded MCP entry

notes-recovery-mcp --case-dir ./output/Notes_Forensics_<run_ts>

Example prompts

  • "Review this NoteStore Lab case root and tell me which artifact to inspect first."
  • "Wire NoteStore Lab MCP to ./output/Notes_Forensics_2026-04-08_... and summarize the review-safe surfaces."
  • "Compare these two case roots and tell me what changed at the manifest/review layer."
  • "Generate a public-safe export plan for this case without exposing raw copied evidence."

Truth language

  • Good: "repo-owned independent skill surface"
  • Good: "local copy-first case review skill"
  • Forbidden: "officially listed skill" without fresh host-side read-back
  • Forbidden: "hosted Notes recovery platform"

Read next

  • references/install-and-mcp.md
  • references/usage-and-proof.md

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

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

补充不同宿主或平台的使用分布数据

能力 5

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

OpenClaw

82.66%
按下载量换算976

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

external-service

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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