[Codex App] Continue an existing ChatGPT conversation in Codex
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 125k
- Forks
- 19.4k
- PR merge metrics
- PR metrics pending
Description
Feature request
Add a first-party Continue in Codex action for an existing ChatGPT conversation.
Desired flow:
ChatGPT conversation → Continue in Codex → choose repository/workspace → resumable Codex thread
Why
Long technical conversations in ChatGPT can contain important context: decisions, logs, screenshots, failed approaches, constraints, and the exact point where work stopped. Today, moving to Codex requires starting a new thread and manually copying or summarizing that context, which wastes time/tokens and can lose useful information.
Possible implementation path
Codex already has an external-agent session import mechanism, used for workflows such as Claude Code → Codex. That seems like a useful precedent for an explicit, user-authorized ChatGPT → Codex import, although ChatGPT-specific privacy or data-model constraints may differ.
The imported conversation does not need to share the same history database; creating a new persistent Codex thread from the relevant ChatGPT context would be enough.
If there is a ChatGPT-specific architectural, privacy, or data-model limitation preventing this, it would be useful to understand what the blocker is.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reviewing Codex's existing external-agent session import mechanism, which the issue identifies as the relevant precedent. Determine how a user-authorized ChatGPT conversation could produce a persistent Codex thread after repository or workspace selection, and document the privacy or data-model constraints; done means the proposed flow is viable or its blocker is clearly established.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- ai
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100