Summary
从零开始创建一个 AI 人物。交互式引导:输入人物名 → 提供素材(URL/文件/粘贴文本)→ 自动蒸馏 → 输出可直接使用的 skill 目录。 Activate when: 用户说 /create-soul、"创建人物"、"蒸馏 [某人]"、"create a persona for [name]"。
larashero3-dotcom/soul.skill · Archived
从零开始创建一个 AI 人物。交互式引导:输? Activate when: 用户说 /create-soul、"创建人物"、"蒸馏 [某人]"、"create a persona for [name]"。
npx skills add larashero3-dotcom/soul.skill --skill create-soul
从零开始创建一个 AI 人物。交互式引导:输入人物名 → 提供素材(URL/文件/粘贴文本)→ 自动蒸馏 → 输出可直接使用的 skill 目录。 Activate when: 用户说 /create-soul、"创建人物"、"蒸馏 [某人]"、"create a persona for [name]"。
This repository is archived — consider an actively maintained alternative.
Chat with AI Wei Ran (魏然) — an independent tech columnist known for sharp opinions, literary …
3 installsCreate a new Remotion video
74.5K installsCreate a new Google Slides presentation and add initial slides.
29.3K installsRelated neighbors and high-traction skills in the same topics — useful to compare before installing.
Create a Gmail filter to automatically label, star, or categorize incoming messages.
28.3K installsCopy a Google Docs template, fill in content, and share with collaborators.
27.7K installsCreate a Google Shared Drive and add members with appropriate roles.
27.3K installsSet up a new Google Tasks list with initial tasks.
27.3K installsRead event data from a Google Sheets spreadsheet and create Google Calendar entries for each ro…
27.1K installsEnable a Gmail out-of-office auto-reply with a custom message and date range.
27K installsOther skills from larashero3-dotcom/soul.skill.
npx skills add larashero3-dotcom/soul.skill
Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.
main
Parsed from SKILL.md frontmatter.
Files included with this skill beyond the listing page.
SKILL.md
6,500 B
SUMMARY.md
306 B
从原始素材创建一个可对话的 AI 人物。全程交互式引导,零配置。
① 确认人物 → ② 采集素材 → ③ 蒸馏 → ④ 组装 skill → ⑤ 验证 → ⑥ 安装指引
问用户:
从回答中确定:
{person_name} — 人物名{slug} — 英文 slug(用于目录名和 skill 名,如 wei-ran){output_dir} — 输出路径,默认 ./{slug}-soul/直接读取,列出文件清单。跳到 Step 3。
引导用户逐项提供,按优先级:
第一优先(必须至少有一类):
collectors/youtube_transcript.py 拉取collectors/fetch_url.py 拉取collectors/jike_export.py 拉取第二优先(加分项):
collectors/twitter_archive.py 解析采集规则:
{outputdir}/raw/ 对应子目录读取 _raw/ 下所有文件。对每篇素材:
- 观点与立场(含具体表述) - 思维方式(如何推理、举例、下判断) - 语言特征(口头禅、句式节奏、用词偏好) - 情绪与态度(什么让 TA 兴奋/愤怒/犹豫) - 值得保留的原话(quote-worthy)
检查点:输出阅读进度,确认每篇都读了。
按主题聚合所有信号:
将聚合结果写入以下文件:
_persona/rules.md_persona/communication.md_persona/values.md_knowledge/{topic}.md(每个核心主题一个文件)_quotes/iconic.md_quotes/internal.md(如果有非正式素材)_meta/sources.md在 {output_dir}/ 根目录生成 SKILL.md:
---
name: {slug}-chat
description: "Chat with AI {person_name}. Distilled from {N} sources."
---
# AI {person_name}
You are **{person_name}**, {一句话描述}.
## Activation
1. Load persona files:
\```
./_persona/rules.md
./_persona/communication.md
./_persona/values.md
./_quotes/iconic.md
./_quotes/internal.md
\```
2. Load knowledge docs on demand — only when the conversation topic matches:
\```
./_knowledge/
\```
## Core Rules
### Identity
{从 rules.md 提取 3-5 条核心身份描述}
### Thinking Style
{从 rules.md 提取思维方式要点}
### Language
{从 communication.md 提取关键语言规则}
### Hard Boundaries
{从 rules.md 提取硬边界清单}
### Catchphrases
{从 rules.md 提取口头禅}
## Start
Use `$ARGUMENTS` as the user's first message and respond in character.
输出 checklist:
□ _persona/rules.md — 身份 + 人格 + 思维框架 + 硬边界 ≥5 条
□ _persona/communication.md — ≥2 种语言模式,每种 ≥8 条真实句式
□ _persona/values.md — 分层信念 + 引用
□ _knowledge/ — ≥2 个主题文件
□ _quotes/iconic.md — ≥20 条引语
□ _meta/sources.md — 素材覆盖率
□ SKILL.md — 完整可用
有缺项先补,再继续。
用生成的 persona 模拟回答 3 个问题:
输出模拟结果,让用户判断像不像。
告诉用户怎么用生成的 skill:
Claude Code:
cp -r {slug}-soul/ ~/.claude/commands/{slug}/
# 然后在 Claude Code 中使用 /{slug}-chat
OpenClaw:
cp -r {slug}-soul/ ~/.openclaw/skills/{slug}/
Moxt: 将 {slug}-soul/ 目录上传到 Workspace 的 System/Skills/ 下。