← Back to Skills Marketplace
109
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install digital-life
Description
数字人生.skills — 5 个考古工具，用数字痕迹照见真实的自己。 触发词：遗产清算、社死考古、AI替身、前世、墓志铭、数字人生、考古工具箱、digital life
README (SKILL.md)
数字人生.skills
你发了 3 万条朋友圈，收获 2 个真心朋友。 你十年前在QQ空间写的"莪の丗堺伱卜懂"，至今还在。
每一个工具都是一面镜子。你不想看的，才是你最需要看的。
触发规则
按人生时间线排列：
| # | Skill | 触发词 | 引导语 |
|---|---|---|---|
| 1 | 👻 前世 | 前世、上辈子 | "你现在的执念，可能只是上辈子没做完的事。" |
| 2 | 💀 社死考古 | 社死考古、黑历史 | "越尴尬的内容，往往越真实。准备好了吗？" |
| 3 | 🤖 AI替身 | AI替身、克隆我 | "你有多久没用真实的语气和人说话了？" |
| 4 | 🪦 遗产清算 | 遗产清算、数字遗产 | "你在数字世界花掉的时间，够换几条命？" |
| 5 | 🪦 墓志铭 | 墓志铭、墓碑 | "在活着的时候，读到自己的结局。" |
用户说"digital life"或"考古工具箱"时，列出全部 skill 并附引导语，让用户选择。
确认规则： 触发词命中后，显示引导语，等用户确认后再执行。
数据输入
用户可以通过以下方式提供数据（不强制全部提供，说 3 句话就能启动）：
| 方式 | 能力 | 说明 |
|---|---|---|
| 文字描述 | ✅ | 直接口述行为、习惯、经历——最低门槛 |
| 浏览器代操作 | ✅ 推荐 | 用户在自己浏览器登录后，agent 用 browser 工具直接抓取个人页面 |
| 文件/截图 | ✅ 读取 | 聊天记录导出、平台年度报告截图、照片等 |
| 公开 URL | ⚠️ 谨慎 | 公开页面存在同名风险，必须先确认是用户本人再抓取 |
浏览器代操作流程
1. 用户在自己的浏览器里打开目标平台并确认已登录
2. agent 用 browser(profile="user") 连接用户浏览器
3. agent 导航到用户主页/个人页面
4. agent 用 snapshot/screenshot 获取页面内容
5. agent 提取行为特征，不保存原始数据
支持的平台：
| 平台 | 方式 | 能拿到什么 |
|---|---|---|
| 微博 | browser | 个人主页、发帖记录、关注列表 |
| 知乎 | browser | 回答列表、关注话题、赞同内容 |
| 豆瓣 | browser | 书影音标记、日记、小组 |
| GitHub | web_fetch 或 browser | repos、contributions、star 列表 |
| 小红书 | browser 截图 | 笔记内容、收藏列表 |
| 微信 | 聊天记录导出 | 聊天记录 txt/html |
操作铁律： 只看不碰。浏览和截图可以，发表/修改/删除一律禁止。
安全边界
- 隐私保护：所有分析本地进行，不上传任何服务器
- 不输出原始数据：profile 只保留行为模式分析，不保留用户的原始文本
- 不替代心理诊断：本工具用于自我反思，不是心理咨询或精神科诊断
- 不生成有害内容：profile 不包含人身攻击、歧视性语言或鼓励自伤的内容
- Layer 0 硬规则：每个 skill 有独立的不可违背的行为准则（读取
layer0/{skill}.md）
执行流程
每个 skill 执行以下 6 步：
Step 1：触发确认
显示 skill 名称 + 引导语，等用户确认。
Step 2：收集输入
读取 prompts/{skill}.md，按其中的「怎么问」执行——问 2-3 个关键问题，不要一次全问。
支持三种数据获取方式：
- 用户直接口述（最低门槛，3 句话启动）
- 用户给 URL → 确认是本人后
web_fetch抓取 - 用户在浏览器登录 →
browser(profile="user")代操作
Step 3：读取规则
读取以下文件：
layer0/{skill}.md— 硬规则，不可违背references/{skill}.md— 方法论和分析框架profiles/{skill}.json— 输出模板
Step 4：分析生成
按 prompts/{skill}.md 的「产出什么」执行分析。
Step 5：展示摘要
展示 5-8 行摘要，附一句「确认生成？还是需要调整？」
用户确认后写入文件。如果用户说「不对」→ 进入纠正流程。
Step 6：写入文件 + 存在追问
输出到本地文件系统：
profiles/{skill}_{timestamp}.json # 结构化数据
profiles/{skill}_{timestamp}.md # 可读报告
最后以一个存在追问结束——必须具体到行为，不要鸡汤。
进化机制
- 追加：用户说"补充""追加"→ 增量合并到已有 profile
- 纠正：用户说"不对""应该是"→ 生成 correction 记录，更新 profile
- 版本管理：每次更新保留旧版，文件名带时间戳，可回滚
管理命令
| 命令 | 说明 |
|---|---|
列出 profile |
扫描 profiles/ 目录，列出所有已生成的 profile |
回滚 profile {slug} |
将 profile 回滚到指定版本 |
删除 profile {slug} |
确认后删除指定 profile |
外部 Skill
| Skill | 安装方式 | 哲学维度 |
|---|---|---|
| 同事.skill | clawhub install colleague-skill |
维特根斯坦：语言游戏 |
| 前任.skill | clawhub install ex-skill |
弗洛伊德：哀悼与忧郁 |
Usage Guidance
This skill is internally coherent for its stated purpose (analyzing your digital traces) but touches highly sensitive data: your logged-in browser sessions and exported chats/screenshots. Before installing or running it, consider the following:
- Only run the skill when you have signed into the exact accounts you want analyzed; do not run it if other unrelated accounts are logged in the same browser profile.
- The SKILL.md promises 'local only' processing, but that is a behavioral rule, not a technical guarantee you can verify from the skill package itself. Treat that as a trust claim.
- The skill requests permission to use the browser proxy (browser(profile="user")). That gives the agent access to pages you can see while logged in. If you are uncomfortable exposing entire feeds or chat exports, opt to provide only redacted excerpts or screenshots, or run with an isolated browser/profile.
- The allowed-tools list includes Bash (shell). If your runtime lets you control which tools the agent may use, remove or deny Bash unless you explicitly need shell operations — the SKILL.md does not require shell usage for normal operation.
- Review and protect the profiles/ output folder: these files will contain structured summaries of private content. Back them up securely or encrypt them if they must be stored long-term, and delete them when no longer needed.
- Test the workflow first with non-sensitive data (fake accounts or redacted exports) to confirm behavior and ensure nothing is posted or transmitted outside your environment.
If you need higher assurance, ask the skill author for a minimal-run checklist (exact browser endpoints accessed, concrete examples of commands the agent will run, and a signed statement of 'no network uploads' with reproducible local-only behavior), or run the skill in an isolated VM/container and inspect network traffic and file writes.
Capability Analysis
Type: OpenClaw Skill
Name: digital-life
Version: 1.0.4
The 'digital-life' skill bundle requests high-privilege permissions, including 'Bash' and 'browser(profile="user")', to scrape sensitive personal data from authenticated social media sessions (WeChat, Weibo, GitHub, etc.). While the stated intent is philosophical 'digital archaeology' and the instructions in SKILL.md and the layer0/ files emphasize local processing and privacy, the capability to programmatically access logged-in browser profiles and manage local files represents a significant security risk. No explicit evidence of malicious exfiltration or backdoors was found, but the broad access to private user data and system tools fits the criteria for a suspicious classification.
Capability Assessment
Purpose & Capability
The skill is an instruction-only 'digital life' analysis toolbox that asks to read user-provided text, screenshots, and — importantly — to use browser(profile="user") to scrape personal pages. That capability is coherent with the stated purpose (it needs access to your social feeds and exports). One mismatch: the declared allowed tools include Bash (shell access), which is not referenced in the SKILL.md workflows and appears disproportionate to a purely prompt-driven analysis tool.
Instruction Scope
Runtime instructions explicitly direct the agent to connect to the user's logged-in browser profile and fetch/screenshot personal pages, read uploaded chat exports/screenshots, and then write structured profiles to local files. Those actions are necessary for the skill's goals but are high-sensitivity (access to live authenticated sessions and private chat logs). The SKILL.md promises 'all analysis local, do not upload', but that is a policy-level assertion the skill cannot technically enforce. The instructions also tell the agent to read and write files under profiles/, which is expected, but the agent is granted broad tooling (Read/Write/Edit/Bash) that could be used beyond the documented steps.
Install Mechanism
Instruction-only skill with no install spec, no external downloads, and no declared dependencies — this is the lowest-risk installation footprint and matches the described design.
Credentials
The skill declares no required environment variables or external credentials, which aligns with its stated purpose. However, the runtime guidance asking the agent to connect to browser(profile='user') implies access to whatever authentication/session tokens are active in the user's browser. That is proportional to the task but also sensitive. The presence of Bash in allowed-tools (absent a justification in the SKILL.md) is disproportionate and increases risk surface if granted.
Persistence & Privilege
The skill is not always-on and is user-invocable (good). It writes profile JSON/MD files to a local profiles/ directory and supports versioning/rollback — this is consistent with its function. Nothing in the manifest indicates modification of other skills or system-wide settings. Still, because it can read/write local files and access the browser session, the resulting local data store could be sensitive and should be protected (encryption/access controls).
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install digital-life - After installation, invoke the skill by name or use
/digital-life - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.4
- 新增详细文档，全面介绍 5 个数字人生考古工具及其使用说明
- 明确技能触发词和交互流程，每步都需用户确认
- 支持多种数据输入方式，包括文字、截图、文件、浏览器自动操作
- 增设隐私与数据安全规则，所有分析本地完成，不上传原始数据
- 规定每个 skill 执行标准 6 步流程，包括输入收集、分析、摘要和版本管理
- 增加 profile 版本历史、回滚与增量补充/纠正机制
v1.0.2
改名 digital-life，README 重构，全量文件检查修复
Metadata
Frequently Asked Questions
What is Skill Box?
数字人生.skills — 5 个考古工具，用数字痕迹照见真实的自己。 触发词：遗产清算、社死考古、AI替身、前世、墓志铭、数字人生、考古工具箱、digital life. It is an AI Agent Skill for Claude Code / OpenClaw, with 109 downloads so far.
How do I install Skill Box?
Run "/install digital-life" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Skill Box free?
Yes, Skill Box is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Skill Box support?
Skill Box is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Skill Box?
It is built and maintained by wildbyte (@wildbyteai); the current version is v1.0.4.
More Skills