clearyss

clearyss / WeLoom

Public

基于本地 RAG 与大模型人格蒸馏的多平台聊天记忆系统。通过 FTS5 检索、语料清洗与长期报告生成,将聊天记录转化为高拟真的数字分身。

10
0
100% credibility
Found May 15, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

WeLoom processes exported chat logs from messaging apps into a local searchable memory system with personality profiles and AI-powered realistic conversations.

How It Works

1
🌟 Discover WeLoom

You hear about a simple way to turn old chat memories into a living digital companion, reviving conversations with loved ones or preserving your own thoughts.

2
📱 Gather Your Chats

Export your favorite chat histories from apps like WeChat or QQ into simple files and place them in a folder.

3
🗂️ Prepare Your Memories

Tell the tool your name from the chats and point it to your files so it knows what belongs to you.

4
🔮 Weave Your Memory Loom

With one command, it cleans, organizes, and builds a searchable treasure chest of your conversations, creating profiles of people and your own style.

5
📊 Generate Insight Reports

Run another quick step to summarize relationships, key events, and your personal traits into easy-to-read stories.

6
💬 Start Heartfelt Chats

Ask questions about past moments or people, and get replies that feel just like the real conversations, full of your true voice.

❤️ Relive Precious Moments

Your digital memories come alive, letting you chat naturally, rediscover forgotten stories, and keep loved ones close forever.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 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 WeLoom?

WeLoom is a Python-based local RAG system that imports JSON chat exports from WeChat or QQ, cleans and indexes them with FTS5 for fast retrieval, then distills a realistic digital twin via LLM persona extraction and long-term report generation. It solves the problem of turning scattered chat histories into queryable, personality-faithful memories—letting you chat with an AI version of yourself or contacts using evidence-based context from your own data. Run CLI commands like `build`, `analyze`, and `chat` for a fully local workflow, with optional OpenAI-compatible APIs.

Why is it gaining traction?

Unlike heavyweight rag github langchain setups or cloud-dependent tools, WeLoom sticks to Python stdlib and SQLite for true local-first RAG github open source—zero frameworks, graceful FTS5 fallbacks, and traceable retrieval keep it lightweight and private. Developers dig the "evidence-first" chats that mimic real speech via style samples and reports, plus easy multi-platform imports without rag github copilot-style vendor lock-in. It's a fresh rag github python example for persona distillation, blooming into a personal memory vault.

Who should use this?

Privacy hawks archiving WeChat/QQ histories for AI-powered recall, indie devs prototyping local rag github repos, or tinkerers building "welcome to your digital self" apps. Ideal for solo creators distilling chat personas without rag github repository bloat, or anyone experimenting with fts5-powered rag github local queries on personal data.

Verdict

Try it if you're into lightweight local RAG projects—solid docs and CLI make onboarding fast, but 10 stars and 1.0% credibility scream early alpha; expect manual tweaks until tests and polish arrive. Great weekend hack for weloom-style memory weaving, not daily driver yet.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.