ZJU-REAL/Easel / skills/builtin/skill-profile-builder
skill-profile-builder
Verified首次使用引导:从社媒链接分析、从零生成账号画像 Profile。收集社媒链接+运营意图,分析已发内容与收藏喜好,生成 6 维 Profile。当用户说 创建画像/第一次用/帮我建个人设/分析我的账号建画像/从链接生成画像 时使用。与 skill-profile-manager 的区别:本 SKILL 专做首次从零分析生成;已有画像的编辑/记忆更新/导出/对比/切换找 profile-manager。
| File | Kind | Size |
|---|---|---|
| EASEL-META.md | markdown | 814 B |
| SKILL.md | skill md | 3.9 KB |
| references/onboarding-questions.md | markdown | 1.9 KB |
| references/social-link-analysis.md | markdown | 1.9 KB |
references/social-link-analysis.md
View on GitHub# 社媒链接分析
从用户提供的主页链接推断画像。目标:把"用户口述"换成"证据支撑",减少主观臆断。
## 能推断什么
| 从公开内容看到 | 推断出的 Profile 维度 |
|----------------|----------------------|
| 近期发帖标题/封面/题材 | identity 内容方向、style 开头结构 |
| 高频出现的语气/表情/句式 | style 语气、标志性元素 |
| 高互动内容 vs 低互动内容 | audience 互动特征、memory 初步洞察 |
| 粉丝量级、更新频率 | platforms、起号阶段判断(供 account-diagnosis) |
| 用户公开的收藏/点赞 | style 审美偏好、想对标的方向 |
| 简介/置顶 | identity 定位、差异化 |
## 抓取方式与降级(重要)
多数国内平台对服务器 IP 有反爬,web_fetch 大概率拿不到完整数据。按以下顺序尝试,**每一步失败就降级,绝不臆造**:
1. **直接 web_fetch 主页链接**(走环境代理)。能拿到就解析近期标题/题材/互动数。
2. 拿不到(403 / 验证码 / 空)→ 告诉用户"该平台主页无法直接抓取",请用户**粘贴**近期几条内容标题、或截图描述、或直接口述常发题材与风格。
3. 仍不足 → 用对标账号 + 用户自述交叉推断,并**标注这是推断、请用户确认**。
> 参考热搜/抓取的通用限制见 `../../../shared/hotlist-apis.md`(同样的反爬规律:不要硬抓平台官网页面)。
## 输出给 Profile 的写法
- 每条推断标来源:`[链接分析]` / `[用户确认]` / `[推断待确认]`。
- 分析结论先跟用户对齐再落盘,避免把模型的猜测写死进 Profile。
- 分析得到的"高互动内容特征"可放进 `memory.md` 作为初步洞察(标注"基于历史内容分析,样本 N 条")。
## 隐私与合规
- 只分析用户**自己**提供的、**公开**的账号数据。
- 不抓取他人隐私、不绕过登录墙、不存储敏感信息。
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |