codeclaw

Export Claude and Codex conversations into privacy-safe datasets for Hugging Face.

11|2|Updated Feb 25, 2026
One-click install
npx skills add https://github.com/ychampion/codeclaw --skill codeclaw-ychampion
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codeclaw
Source: https://github.com/ychampion/codeclaw/tree/main/docs
Command: npx skills add https://github.com/ychampion/codeclaw --skill codeclaw-ychampion

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires huggingface_hub, pyyaml, json, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Empower users to export Claude Code and Codex conversation histories as organized, privacy-safe training data.

Core Features & Use Cases

  • Data Conversion: Converts chat sessions into prepared datasets suitable for machine learning or analysis.
  • Privacy Management: Provides redaction, anonymization, and review mechanisms to safeguard sensitive information.
  • Use Case: A developer wants to share anonymized conversation logs with colleagues or upload datasets to Hugging Face without exposing PII.

Quick Start

Run the codeclaw export command to generate a dataset from your conversations and review the output before pushing it to Hugging Face.

Frequently Asked Questions about codeclaw

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I export Claude conversation logs into a structured dataset?

You can export Claude conversation logs into a structured dataset by running the export command, which converts chat sessions into prepared data formatted for machine learning analysis and training resource sharing.

Can I redact personal information before uploading conversation datasets to Hugging Face?

Yes, you can redact personal information before uploading to Hugging Face using built-in privacy management mechanisms. These scripts provide anonymization and review features to safeguard sensitive data and ensure PII is not exposed in shared datasets.

Does codeclaw support exporting conversations from both Claude and Codex models?

Yes, codeclaw supports exporting conversation histories from both Claude Code and Codex models. It facilitates managing this conversation data uniformly, ensuring privacy compliance and easy dataset sharing across different AI model platforms.

What is the best way to format chat histories for machine learning training data?

The best way to format chat histories for machine learning training data is using Python scripts that convert coding conversations into organized, privacy-safe datasets. This approach ensures structured formatting suitable for AI development and research teams.

Do I need huggingface_hub to share my exported conversation datasets?

Yes, you need the huggingface_hub dependency to push and share your exported conversation datasets. The Skill utilizes this library alongside pyyaml and json to facilitate uploading formatted data to the Hugging Face platform.