A professional-grade skill designed to parse, analyze, and transform raw ChatGPT JSON exports into a searchable, structured personal memory archive.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install chatgpt-memory-extraction
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install chatgpt-memory-extraction using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is ChatGPT Memory Extraction?
The ChatGPT Memory Extraction tool is an essential addition to the Openclaw Skills ecosystem for users seeking to reclaim their digital history. It goes beyond simple keyword searching by deeply reading every interaction within a ChatGPT export to identify recurring themes, significant life events, and evolving ideas.
By processing the standard conversations.json file, this skill synthesizes fragmented dialogues into a coherent personal diary and knowledge base. It is specifically engineered to handle the technical nuances of OpenAI's data exports, ensuring that no valuable insight is lost during the migration from a transient chat interface to a permanent, structured format.
ChatGPT Memory Extraction Use Cases
- Creating a chronological timeline of personal milestones and projects discussed with AI.
- Organizing a personal knowledge management (PKM) system based on past research and brainstorming sessions.
- Building detailed people profiles for individuals mentioned throughout years of conversation history.
- Migrating long-term memory and context from ChatGPT to local, self-hosted AI systems.
- Conducting a thematic review of personal growth or technical learning over specific time periods.
How ChatGPT Memory Extraction Works
- Data Acquisition: The user initiates a formal data export via the ChatGPT settings and downloads the resulting ZIP archive containing the conversations JSON files.
- Parsing and Indexing: The skill utilizes a specialized extraction script to convert raw, nested JSON data into a series of human-readable text files and a comprehensive conversation index.
- Batch Processing: To maintain high accuracy and prevent AI context limits, the skill processes history in quarterly or monthly batches, reading every conversation in full.
- Dimensional Extraction: As the skill reads, it simultaneously updates specialized files for different dimensions, such as people, specific topics, and emotional arcs.
- Incremental Synchronization: The system tracks conversation IDs to allow for future incremental updates without re-processing the entire historical archive.
ChatGPT Memory Extraction Setup
To begin using this tool within the Openclaw Skills framework, ensure you have your conversations.json file ready in your working directory.
# Navigate to the skill directory
cd skills/chatgpt-memory-extraction
# Run the extraction script to convert JSON to readable text
python scripts/extract_conversations.py --input path/to/conversations.json --output ./extracted_data
Follow the on-screen prompts to begin the quarterly processing and timeline generation.
ChatGPT Memory Extraction Data Schema & Taxonomy
The skill organizes extracted data into a strict taxonomy to ensure maximum searchability and clarity:
| Component | Description | Format |
|---|---|---|
| Timeline | Chronological logs of all significant events and discussions. | Markdown per month/quarter |
| People Profiles | Dedicated files for every individual mentioned in conversations. | Individual .md files |
| Thematic Records | Subject-specific archives (e.g., Coding, Philosophy, Travel). | Categorized Markdown |
| Conversation Index | A master list mapping JSON IDs to local file summaries. | JSON/Markdown Table |
| Metadata | Timestamps, sentiment tags, and interaction frequency. | YAML Frontmatter |
ChatGPT Memory Extraction Advanced Features
- Support for incremental processing to merge new OpenAI exports with existing archives seamlessly.
- Multi-dimensional extraction that automatically cross-references people and topics across different conversations.
- Batch-aware processing logic that splits high-volume history (100+ chats) into manageable monthly segments.
- Integration with local quality rules to mitigate common AI hallucinations or summarization errors.
- Automated generation of thematic summaries for long-term usage reviews.
SKILL.md
Loading
Related Openclaw Skills
Cloud Architecture Holistic View
A comprehensive cloud management skill for visualizing architecture diagrams, evaluating risks, and managing Tencent Cloud Smart Advisor resources via AK/SK authentication.
Test-Test Skill
A specialized diagnostic skill designed to validate the integration and responsiveness of AI coding agents.
Video Transcription and Content Optimization
An automated workflow that transforms raw video/audio files into high-quality transcriptions and platform-specific articles with professional polishing.
Deputy
A comprehensive workforce management skill that integrates Deputy with AI agents to automate scheduling, time tracking, and employee data management.
Wudao A-Share Limit-Up Analysis
A professional-grade market analysis tool for tracking and analyzing A-Share limit-up stocks, board ladders, and real-time trading sentiment.
Walmart Store Analyzer
An automated Walmart marketplace intelligence agent that transforms seller traffic reports and item data into prioritized growth strategies.
Featured*
