okhp3-chatgpt-project-migration

Migrates ChatGPT project exports into hashed, reconciled repository artifacts with provenance tracking.

1|Updated Jun 13, 2026
One-click install
npx skills add https://github.com/OKHP3/refoldec --skill okhp3-chatgpt-project-migration-okhp3
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: okhp3-chatgpt-project-migration
Source: https://github.com/OKHP3/refoldec/tree/main/.agents/skills/okhp3-chatgpt-project-migration
Command: npx skills add https://github.com/OKHP3/refoldec --skill okhp3-chatgpt-project-migration-okhp3

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Migrating ChatGPT project threads into a repository often loses regenerated branches, attachments, and provenance, making it impossible to prove what was captured or to safely publish distilled material. This Skill enforces a lossless capture-first workflow that preserves raw exports, normalizes every conversation node, and reconciles threads before any artifact is committed. ## Core Features & Use Cases - Lossless source preservation: Archives the original export ZIP outside Git with SHA-256 hashes, keeping raw evidence separate from public-safe artifacts. - Branch-complete normalization: Parses every node in the conversation mapping object, including regenerated alternatives and non-current branches, rather than only the visible current_node path. - Reconciliation and exception tracking: Matches UI-inventoried threads to export records by conversation ID and keeps unmatched threads, branches, and assets as open exceptions until resolved. - Use Case: You have a 200-thread ChatGPT project to fold into your repo. Use this Skill to inventory every thread, hash the export, normalize all mapping nodes, recover unmatched threads via browser capture, and route only reviewed, public-safe distillations into committed documentation. ## Quick Start Ask the agent to capture and migrate your ChatGPT project into the repository using the okhp3-chatgpt-project-migration workflow, starting with a project capture manifest.

Frequently Asked Questions about okhp3-chatgpt-project-migration

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

FAQPage Schema
How do I migrate ChatGPT project conversations into a Git repository?▼

Export the ChatGPT data, archive the raw ZIP outside the repository with a recorded SHA-256 hash, then inventory every thread by conversation ID in a capture manifest. Normalize each conversation, reconcile against the UI inventory, and commit only reviewed, public-safe artifacts.

How to preserve regenerated ChatGPT responses when parsing an export?▼

Retain every key in the conversation mapping object, not just the current_node path. Each regenerated response is a sibling branch in the mapping, so normalized records must keep node_id, parent_id, and child_ids for all nodes.

What should I do when a ChatGPT thread has no export match?▼

Assign the thread a no_match status and open an exception rather than dropping it. Use browser-based capture as the recovery path, recording the URL, capture time, and reason, and keep the exception open until recovered or accepted as a loss.

Can I commit raw ChatGPT conversation exports to a public repository?▼

No. Raw exports are treated as immutable evidence kept in a local archive outside Git. Only reviewed, classified, public-safe Markdown distillations with source references should be committed to repository paths.

What are the limitations of UI-only ChatGPT capture without a data export?▼

UI-only capture is marked as ui_capture_only with reduced fidelity, since it may miss hidden branches, metadata, and attachments available in a full export. The batch must list these limitations and treat the export as the primary source whenever available.