Claude Codex Consult — AI skill for Claude Code
Claude Code skill + PowerShell bridge that lets Claude consult OpenAI Codex (any model) mid-task over your ChatGPT subscription: fork/resume threads, read-only sandbox, file ledger of every consultati.
How to install Claude Codex Consult
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open xelth-com/claude-codex-consult and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Claude Codex Consult does
Claude Code skill + PowerShell bridge that lets Claude consult OpenAI Codex (any model) mid-task over your ChatGPT subscription: fork/resume threads, read-only sandbox, file ledger of every consultation
Alternatives in AI
- Surface — Show ranked attack surface for a target from its recon manifest + hunt memory 3.8k ★
- Multi Model Consult — Multi-Model Consultation Skill 1.9k ★
- Azure AI Agents .net — Persistent AI agents with threads and tools 1.8k ★
README
claude-codex-consult
Claude Code coordinates; **Codex is a standing reasoning partner you can call mid-task**. This plugin adds one skill and one small PowerShell bridge so Claude can hand Codex a one-page brief and get a judgement back — over your existing ChatGPT subscription, with no API key. The same `` keeps the Codex thread alive across consultations (`fork` to branch, `resume` to continue), Codex runs in a **read-only sandbox** by default, and every consultation lands on disk as a file pair — your brief and Codex's verbatim reply — plus a JSON ledger entry. Consultations become reviewable history you can commit next to the code they were about, instead of chat you lose.
Any Codex model works. Leave the model unset and Codex uses whatever is in your `~/.codex/config.toml`; pass `-Model ` when you want a specific one.
Install
/plugin marketplace add xelth-com/claude-codex-consult
/plugin install codex-consult@claude-codex-consult
Then just ask Claude to consult Codex, or invoke the skill directly:
/codex-consult:consult-codex my-task should we cache at the edge or in the worker?
Requirements
| Claude Code | any recent version with plugin support |
| Codex CLI | ≥ 0.148 — codex exec fork landed there. Developed against 0.155.1 |
| Codex auth | signed in with your ChatGPT account (codex login). No API key needed |
| Shell | Windows PowerShell 5.1 (preinstalled on Windows) or PowerShell 7 (pwsh) on macOS/Linux/Windows |
| git | optional — used to locate the project root and stamp the reviewed revision |
Usage
**1. Claude writes a brief** to `.collab//handoffs/-claude-.md`: question, task state, evidence with `file:line`, the alternatives already weighed, numbered questions Q1…Qn, and a word cap. One page.
**2. One command** (Claude runs this for you):
powershell -NoProfile -ExecutionPolicy Bypass -File "${C
Related Skills
Cobol Legacy Ledger
Learn COBOL through a live banking system — 18 programs, 6-node settlement network, real-time simulation, Pyth
Ayontclaudian
ayontclaudian — Obsidian plugin embedding Claude Code, Codex, Antigravity & Kimi coding-agent CLIs with a unif
Windows DNS Policy Manager
Browser-based GUI for Windows Server DNS — policies, zones, records, DNSSEC, DNS over HTTPS, and blocklists. N
Ollama Cc Plugin
Use local/cloud ollama models from Claude Code as a second engine: read-only consult/review, agentic worktree-
Mirofish CLI
Multi-agent AI prediction engine - digital sandbox for scenario simulation (fork of 666ghj/MiroFish)
Agenvoy
Self-hosted AI agent harness in a single Go binary — writes, sandbox-tests and repairs its own tools, and lets
Related Agents
Codex Agent
Runs OpenAI Codex on the user's ChatGPT subscription. Three modes - consult (OS-sandboxed read-only analysis,
Tsk - AI Agent Task Manager And Sandbox
by dtormoen - A Rust CLI tool that lets you delegate development tasks to AI agents running in sandboxed Docke
GPT 5.6 Luna
General-purpose agent that runs on GPT-5.6-Luna (routed through the shunt gateway to the ChatGPT/Codex subscri