NatalieCao323

NatalieCao323 / crush-skill

Public

crush.skill — 贝叶斯暧昧分析引擎 | A Bayesian memory engine for your situationship

19
0
100% credibility
Found Apr 05, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

An AI skill that processes chat logs and photos from messaging apps to analyze romantic interest, provide diagnostic reports, and simulate conversations for personal clarity.

How It Works

1
🥺 Wondering about your crush

You're getting mixed signals from someone you like and want to know if they feel the same.

2
💡 Discover crush.skill

You find this helpful tool online that promises clear insights from your chat history.

3
🛠️ Set it up quickly

Follow simple steps to get it ready on your computer so you can start using it right away.

4
📝 Create a profile

Tell it a bit about your crush and prepare your chat logs from apps like WeChat or QQ.

5
✨ Upload and analyze

Share your messages and photos – it reads them all and tags the important emotional clues instantly.

6
📊 Read the report

Get a score on how things are progressing, their attachment style, and exact scripts for what to say next.

7
💬 Practice conversations

Chat with an AI that texts just like them to rehearse your next messages safely.

❤️ Gain total clarity

You now know if it's real or time to move on, feeling empowered and at peace.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 19 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is crush-skill?

crush.skill is a Python-based Bayesian memory engine that analyzes your chat logs from WeChat, QQ, or plain text to decode situationships—those ambiguous romantic connections. Feed in messages or photos, and it tags them with prior confidence, time decay, and emotional intensity, spitting out progression scores, attachment style diagnoses, psychological breakdowns, and next-move scripts. Users also get a simulation mode that mimics your crush's texting style for rehearsal.

Why is it gaining traction?

It stands out with local-only processing for total privacy—no data leaves your machine—plus seamless integration as a skill for Claude Code and OpenClaw via simple CLI installs like `openclaw skills install crush`. The hook is its crush cli skills for real-time diagnostics, like Bayesian progression scores that cut through "maybe next time" vagueness, blending crush agent skills with memory engine smarts in a way generic chat analyzers don't.

Who should use this?

AI skill builders on Claude or OpenClaw tweaking personal agent workflows, or Python devs experimenting with Bayesian engines for sentiment tools. Ideal for anyone parsing WeChat/QQ exports to analyze crush skills in situationships, from photo timelines to response simulations—perfect if you're scripting your own emotional radar.

Verdict

With 19 stars and a 1.0% credibility score, it's raw and early-stage—docs are solid in English/Chinese but lacks tests or broad adoption. Worth forking for Bayesian crush analysis experiments, but productionize it yourself first.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.