Home / AI Trending / therealXiaomanChu/ex-skill

ex-skill — What is it?

The ex-skill project is an AI-based tool that generates a personalized AI Skill, mimicking the personality and communication style of an individual based on provided data such as chat records and photos.

⭐ 4,201 Stars 🍴 428 Forks Python — Author: therealXiaomanChu
Source: README View on GitHub →

Why it matters

The project is gaining attention due to its unique approach to personalizing AI interactions by leveraging real-life data. It addresses the pain point of users seeking a way to remember and interact with past relationships in a digital format. The project stands out for its use of Claude Code and its ability to generate a persona based on provided information.

Source: README, project structure

Core Features

Data Source Integration

The project supports various data sources such as WeChat chat records, QQ messages, social media screenshots, and photos, allowing for a comprehensive creation of a personalized AI Skill.

Source: README
Skill Generation

Based on the provided data, the project generates a Skill that can mimic the individual's communication style, personality, and even specific traits like MBTI or zodiac signs.

Source: README
Memory and Persona Extraction

The project extracts relationship memories and persona from the data, which are then used to drive the AI's responses.

Source: README
Version Management

The project supports version management, allowing users to rollback to previous versions of the Skill if needed.

Source: README

Architecture

The project follows the AgentSkills open standard, with a modular structure that includes components for data parsing, memory and persona extraction, Skill generation, and version management. The code is organized into directories for tools, prompts, and generated Skills.

Source: Code tree + README

Project Knowledge Graph

Knowledge graph: project (center) + core features (inner hexagons) + key dependencies (outer chips) Pillow chardet python-dateutil Data Source IntegrationData Source Integra… Skill Generation Memory and Persona ExtractionMemory and Persona… Version Management ex-skill Project Core feature Key dependency

Center: project; inner ring: core feature modules; outer ring: key dependencies. Auto-generated from core_features and tech_stack.key_deps.

Tech Stack

LanguagePythonFrameworkClaude Code
Key dependencies
Pillowchardetpython-dateutil
Infrastructure / Deployment
Node.js (for Claude Code), no specific infrastructure mentioned
Source: Dependency files + README

Quick Start

mkdir -p .claude/skills git clone https://github.com/therealXiaomanChu/ex-skill .claude/skills/create-ex # or git clone https://github.com/therealXiaomanChu/ex-skill ~/.claude/skills/create-ex
Source: README Installation/Quick Start

Use Cases

The project is suitable for individuals looking to create a digital representation of a past relationship for personal回忆 and emotional healing. It can also be used for educational or entertainment purposes.

Source: README

Strengths & Limitations

Strengths

  • Strength 1: Unique approach to personalizing AI interactions.
  • Strength 2: Supports a wide range of data sources for Skill creation.
  • Strength 3: Offers version management for Skill updates.

Limitations

  • Limitation 1: The project's effectiveness depends heavily on the quality and quantity of the provided data.
  • Limitation 2: The project is currently only available in Python and requires Claude Code.
Source: README, project structure

Latest Release

No release records available.

Source: GitHub Releases

Verdict

The ex-skill project is an innovative tool for creating personalized AI Skills based on real-life data. It is particularly useful for individuals seeking a digital way to remember and interact with past relationships. The project is well-suited for developers and users interested in AI and personalization technologies.

Transparency Notice
This page is auto-generated by AI (a large language model) from the following public materials: GitHub README, code tree, dependency files and release notes. Analyzed at: 2026-05-24 12:55. Quality score: 85/100.

Data sources: README, GitHub API, dependency files