Credentials. Claude mode uses your authenticated Claude Code client, so keep your subscription login and any worker route credentials under your own control.
Who it is forA developer with an Opus route and a compatible native worker who wants a stronger planner and a cheaper implementer on the same build.
The problemUsing one model for both planning and implementation may not fit the work, and manually moving context between models is error-prone.
The outcomeOpus produces the plan, Codex dispatches the implementation to the native worker, and the result returns to Opus and Codex for review.
What you need first
- Node.js 24.15 or newer
- A Codex client with plugin support
- Model access you already have, and a compatible configured route for native workers
- The official Claude Code CLI, authenticated through your subscription, for Claude mode
Steps
- Run
codex plugin marketplace add "$PWD"from this checkout in a stable folder. - Run
codex plugin add codex-on-crack@codex-on-crack. - Read the setup guide and configure a compatible native worker with the crack-setup skill, reviewing the preview before applying it.
- Start a fresh conversation and ask Opus to plan the feature and Flash to implement it, then bring the result back for review.
- Check the returned worker results and host-tool evidence before accepting the work.
Prompt for your agent
Use the codex-on-crack package in <project folder> to have Opus plan <feature name> and a configured native worker implement it, then return the result to Codex for review. Expected output: a plan, the worker's changes, and a review note from the lead. Verify that the review cites the worker results and host-tool evidence.
Caveat. Instructions and file scopes are not an OS sandbox, so the worker can still act within your permissions.
Get 5 new use cases every morning. Free, one short email, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Disclaimer
Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page