Eddiearc Codex Delegator
eddiearc-codex-delegator is a Claude Code plugin for the Build phase that adds an Agent Skill to delegate complex logic tasks to OpenAI Codex CLI.
by eddiearc · github.com/eddiearc/codex-delegator
Install when Claude Code should hand off heavy logic or reasoning work to OpenAI Codex CLI instead of forcing one model to do everything inline.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install eddiearc-codex-delegator@eddiearc/codex-delegatorBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:eddiearc/codex-delegator") and the paste-ready config with skillselion.get_install_config("plugin:eddiearc/codex-delegator").
What it does
eddiearc-codex-delegator is a Claude Code plugin built around one Agent Skill that delegates demanding, logic-heavy tasks to the OpenAI Codex CLI. Solo builders who already run Claude Code for implementation but hit walls on intricate algorithms, gnarly refactors, or deep reasoning chains can use it to split labor: Claude orchestrates the repo context while Codex tackles the intensive slice. The value proposition is collaboration between agents rather than a larger prompt in a single session. You should plan for two toolchains—Claude Code plus a working Codex CLI setup with appropriate credentials—and accept slightly more moving parts in exchange for better outcomes on hard problems. It fits the Build phase agent-tooling shelf because it changes how work is executed, not how you validate market fit or operate production monitors. Beginners can try it on bounded tasks; advanced users gain most when they define clear delegation boundaries so context passes cleanly between agents.
Highlights
- Single Claude Code Agent Skill focused on delegating logic-intensive work
- Automatic handoff pattern to OpenAI Codex CLI for AI-to-AI collaboration
- Aimed at complex problem-solving that benefits from a second reasoning engine
- Catalog-listed under skills-style plugin metadata with codex and delegator keywords
- Lightweight one-plugin install versus a full multi-skill framework
Why builders use it
One Claude Code session often stalls on dense logic while you still pay latency and context cost trying to brute-force the same model.
After install, Claude Code can automatically delegate selected complex, logic-intensive work to Codex CLI and merge results back into your workflow.
At a glance
- Type - Plugin in AI Agents.
- Adoption - 0 installs, 13 stars, 0 votes.
FAQ
Who is eddiearc-codex-delegator for?
It is for Claude Code users who also run OpenAI Codex CLI and want automatic delegation for logic-heavy tasks.
When should I use eddiearc-codex-delegator?
Use it when a task is too logic-intensive for a single inline pass and splitting work to Codex is likely to improve quality or speed.
How do I add eddiearc-codex-delegator to my agent?
Install the eddiearc/codex-delegator Claude Code plugin and ensure OpenAI Codex CLI is installed, authenticated, and reachable from the environment Claude uses.
Comments
Share how you use eddiearc-codex-delegator, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.