darkdevil2806

Shared Chat Extractor — Communication skill for Claude Code

Communication community

Fast, zero-dependency CLI tool & Antigravity skill to extract conversations, metadata, and code snippets from ChatGPT, Claude, Gemini, and Perplexity share links into clean Markdown and structured JSO.

How to install Shared Chat Extractor

This entry records only its repository, not the path inside it, so there is no exact command to give. Open darkdevil2806/shared-chat-extractor and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Shared Chat Extractor does

Fast, zero-dependency CLI tool & Antigravity skill to extract conversations, metadata, and code snippets from ChatGPT, Claude, Gemini, and Perplexity share links into clean Markdown and structured JSON.

Alternatives in Communication

  • Prompts.chat — f.k.a. Awesome ChatGPT Prompts 154.3k ★
  • Post Longform — Publish a longform video to YouTube with title, description, and tags via the creatoros CLI 212 ★
  • Dsh Chat Import — Import 14+ external agent chat histories (Claude Code, Codex, ChatGPT, Cursor, Gemini, Reasonix, opencode, ZCo 114 ★

README

Python 3.7+ Zero Dependencies MIT License Supported Platforms

🗂️ Shared Chat Extractor

A fast, lightweight, dependency-free tool to extract structured conversation turns, metadata, and code snippets from **ChatGPT**, **Claude**, **Google Gemini**, **Perplexity**, and other AI share links or offline HTML pages.

Converts AI conversation links into clean **Markdown**, structured **JSON**, or standalone **source code files**. Runs as a standalone CLI tool **and** as an official [Google Antigravity (AGY)](https://github.com/google/antigravity) skill.


✨ Key Features

Feature Description
🔗 Direct URL Extraction Paste a share link to automatically fetch and parse turns & metadata
💾 Automatic HTML Saving Save the raw fetched .html payload to disk with --save-html
🌐 Offline DOM Parsing Parse local .html files saved from browser sessions (Cmd+S / Ctrl+S)
📝 Markdown Output Clean, human-readable Markdown with role headers and formatting
📊 Structured JSON Full JSON with roles, model identifiers, timestamps, and metadata
💻 Code Extraction Automatically extract all code snippets into separate files (.py, .js, etc.)
🔄 Multi-Architecture Support Supports React Router turbo-stream, __NEXT_DATA__, OpenGraph negotiation, and DOM queries
🔒 Zero Dependencies Pure Python standard library (urllib, re, json, ssl) — nothing to install

📦 Installation

No package installations or virtual environments required:

git clone https://github.c