Route Codex and Claude Code through an opencodex provider proxy
Run a local proxy that lets Codex and Claude Code use configured LLM providers, routed models, and account pools while keeping the normal coding-agent workflow.
npx skills add agentskillexchange/skills --skill route-codex-and-claude-code-through-an-opencodex-provider-proxy
npm install -g @bitkyc08/opencodex, run ocx init, start the proxy with ocx start, optionally install the Codex shim with ocx codex-shim install, then configure providers through ocx gui or the generated config.Use opencodex when an operator wants Codex CLI, Codex App, Codex SDK, or Claude Code to keep their normal workflow while routing model calls through a local provider proxy. The workflow is: install the package, run `ocx init`, configure provider credentials or OAuth in the dashboard, start the proxy or install the service/shim, then use `codex` or `ocx claude` with provider/model routing. Invoke this instead of using a provider product normally when the work should stay inside the existing coding-agent runtime but needs Anthropic, Gemini, Grok, DeepSeek, Ollama, OpenRouter, Azure, or another OpenAI-compatible backend. The scope boundary is local model-routing and provider/account management for Codex and Claude Code sessions. It is not a generic LLM catalog, hosted model platform, SDK reference, or replacement coding agent.
What this skill actually does
Inputs and prerequisites: Node.js 18+, @bitkyc08/opencodex, Codex CLI/App/SDK or Claude Code, provider credentials or OAuth.
Setup notes: Install with npm install -g @bitkyc08/opencodex , run ocx init , start the proxy with ocx start , optionally install the Codex shim with ocx codex-shim install , then configure providers through ocx gui or the generated config.
Source and verification boundary: use https://lidge-jun.github.io/opencodex/ as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.