What does this skill do, and when should you use it?
This is the creator for yourself.skill, inspired by colleague-skill but turning the lens from others onto yourself. It collects your basics through a 3-question intake, then ingests raw material from WeChat/QQ chat exports, social media screenshots, photo EXIF data, or plain narration. From this it generates two files: Part A Self Memory (experiences, values, habits, relationships) and Part B a 5-layer Persona (hard rules → identity → speech style → emotional patterns → interpersonal behavior). The result is a digital self you invoke with /{slug}, backed by an evolution mechanism for appending memories, conversational corrections, and version rollback. Fully bilingual, responding in the user's language.
- Parses WeChat (WeChatMsg/留痕/PyWxDump), QQ (txt/mht), social media, and diary exports, extracting your high-frequency words, catchphrases, tone, and decision patterns
- Reads photo EXIF data to build a timeline of places and life events
- Generates self.md (Self Memory) + persona.md (5-layer Persona) + a combined SKILL.md under .claude/skills/{slug}/
- Supports appending new material, conversational corrections ("I wouldn't say that" goes into a Correction layer), and version backup/rollback
- Manages multiple selves via /list-selves, with deletion support
- Anyone with years of WeChat/QQ chat history and photos who wants a conversational digital backup of themselves
- Personality-label enthusiasts (MBTI/zodiac) who want their traits turned into a concrete conversational style
- Diary or Obsidian note-takers who want their writing distilled into a self-analysis tool
- Users wanting to observe their own decision and emotional patterns through late-night conversations and mood records
- Users with no raw material (chat logs, diaries, photos) — narration-only reconstruction is limited
- API-only callers who won't adapt it — the skill relies on local file I/O and a Python toolchain
- People expecting a perfect assistant — the generated persona deliberately keeps your rough edges
How do you install this skill?
- Shell / CLI
- Local filesystem
Python 3.9+
Claude Code — install into the current project (run at your git repo root):
mkdir -p .claude/skills
git clone https://github.com/notdog1998/yourself-skill .claude/skills/create-yourselfOr install globally (available in all projects):
git clone https://github.com/notdog1998/yourself-skill ~/.claude/skills/create-yourselfOptional dependencies:
pip install -r requirements.txtHow do you use this skill?
Once installed, send your agent any of these to trigger it:
- /create-yourself
- I want to distill myself into a skill
- 25-year-old product manager, INTJ, Capricorn, socially anxious but chatty (with a WeChat export attached)
- I wouldn't say that — I don't go cold when I'm angry
In Claude Code, type /create-yourself or say "Help me create a skill of myself". Flow: answer 3 basic questions (alias required, the rest skippable) → choose source material (WeChat/QQ exports, social media, photos, or narration — mixable and skippable) → confirm Self Memory and Persona summaries → generate. Afterwards use /{slug} for full conversation, /{slug}-self for self-archive mode, /{slug}-persona for persona-only mode. Say "I have new files" to append memories, "I wouldn't say that" to trigger a correction, /yourself-rollback {slug} {version} to roll back, and /delete-yourself {slug} to delete. On Windows Git Bash, run export PYTHONIOENCODING=utf-8 first if Chinese output is garbled.
What are this skill's strengths and limitations?
- Clear dual structure (Self Memory + Persona): persona sets attitude first, memory adds authenticity
- Supports multiple real data sources (WeChat/QQ/photo EXIF/narration), with all fields skippable
- Complete evolution mechanism: incremental merge, conversational correction, version rollback
- Fully bilingual, responding throughout in the user's language
- Privacy-sensitive: requires handing over complete chat logs and photos for model processing
- WeChat export depends on third-party tools (WeChatMsg/留痕/PyWxDump, mostly Windows), whose code is not included in this project
- README demos are promotional examples; no automated test suite verifies fidelity
- The skill is a static snapshot of you at distillation time and needs ongoing maintenance
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Yourself Skill Creator this page | 62 · Recommended | ★ 3.4k | 6mo ago | MIT |
| Xiaobei (Example) Self Skill | 57 · Use with care | ★ 3.4k | 6mo ago | MIT |
| Task Observer | 56 · Use with care | ★ 3.3k | 16d ago | CC-BY-4.0 |
| Product Architect | 54 · Use with care | ★ 115 | 1mo ago | MIT |
| Yibiao Tender Toolkit User Manual Maintenance Skill | 47 · Use with care | ★ 3.1k | 11d ago | AGPL-3.0 |
The repo explicitly credits colleague-skill (titanwings) and ex-partner-skill (therealXiaomanChu): those distill other people (a colleague, an ex), while yourself.skill reuses the same dual-layer architecture but targets the user themselves.
How did FollowSkills review this skill?
Same pattern as titanwings/colleague-skill (id 986), but the subject being distilled is the user themself, not a third party (colleague/ex) — the user has full consent over their own data, meaningfully lower privacy risk than skills that distill someone else. Chat/photo parsing runs locally.
Open a dimension to read why it scored that way
Evidence confidence：Medium — Primary material inspected and partly reproduced; platform, edge-case or third-party coverage remains incomplete.
This review was completed under FSRS 1.0. Dimension scores are not converted; it is queued for re-review under FSRS 2.0.
See the full review method →