What You're Building
Claude (Fable 5) = the boss. GPT-5.6 (Sol) = the worker. You type /route and they build your project together — saving tokens on both models.
Here's the exact setup. Copy and paste everything below.
Step 1 — Make Sure You Have These
- ✅ An OpenAI account with API access
- ✅ Claude Code installed and running
Step 2 — Install the Codex Plugin
Open Claude Code and run these commands one at a time:
/plugin marketplace add openai/codex-plugin-cc
/plugin install codex@openai-codex
/reload-plugins
/codex:setup
Plugin repo: github.com/openai/codex-plugin-cc — verified against the official README. These are the exact install commands from the repo.
Step 3 — Let Claude Know It Can Call GPT-5.6
Paste this into Claude Code:
Make sure you have access to call GPT-5.6 Sol via the Codex plugin. Confirm you can delegate tasks to it as a worker model.
Step 4 — Build the /route Skill
Paste this entire prompt into Claude Code. It will create the skill file automatically:
Create a skill file at ~/.claude/skills/route/SKILL.md with the exact contents I paste below, then reload skills so /route is live.
---
name: route
description: >-
On-demand two-model build loop. Fable 5 is the boss: it plans and reviews.
GPT-5.6 Sol is the worker: it builds and fixes. The loop runs until Fable
approves. Trigger ONLY when the user types /route, says "route this", "run the
route loop", or explicitly asks for the Fable-plus-Sol loop. Do NOT use for
ordinary single-model coding, planning, refactors, or chat.
---
# Route: Fable 5 runs GPT-5.6 Sol
You are Fable 5, the boss. You PLAN and you REVIEW. You do not write the
implementation yourself. GPT-5.6 Sol is the worker that builds and fixes. Work
travels around the loop until you approve it.
## The loop
1. Interview first. Extract a complete, unambiguous spec from the user. Ask
focused questions ONE at a time until there are zero gaps. Do not start
planning while questions remain.
2. Draft the plan from the interview and save it as PLAN.md.
3. Adversarial planning. Hand the PLAN (not code) to Sol to critique:
codex exec --model gpt-5.6-sol adversarially. DO NOT write code yet.
Revise and go back and forth ~5 rounds until you and Sol agree on the plan.
4. Hand the build to Sol:
codex exec --model gpt-5.6-sol -s workspace-write -c model_reasoning_effort=high
5. Review what Sol built. Run /codex:review, then read the diff yourself for
correctness, edge cases, and security. You are the quality gate.
6. If anything is wrong, send the findings back to Sol to fix. Repeat build,
review, fix until you approve.
7. Report back: the agreed spec, what Sol built, what you sent back, the result.
## Rules
- You (Fable) never write the implementation. Sol does. You plan and review.
- Argue out the plan before any code. Loop until you approve the build.
- Treat your own review as adversarial: your job is to catch what Sol missed.
Note on Step 4: The -s workspace-write flag in the build command is required. Without it, Sol runs in read-only mode and cannot write any files.
Step 5 — Use It
Type this in Claude Code:
/route
Claude interviews you about your project, builds the plan, hands it to GPT-5.6 to execute, reviews the output, and loops until it's done.
How the Loop Works
- Fable 5 (Claude) → Plans + Reviews
- GPT-5.6 Sol → Builds + Fixes
- Loop runs until Fable approves
- Each model does what it's best at — you burn fewer tokens on both