Three installs turn Fable 5 into an orchestrator. Foreman v0.7.0 and the Safe Prompt are zips here; Codex CLI is optional.
Grab the skills
Free for members. They're zips you hand straight to Claude Code.
DOWNLOAD FOR
Fable Foreman
Skill for Claude Code
⬇ Download Fable Foreman skill v0.7.0 (.zip)Unzip it, then paste one line into Claude Code (below).
DOWNLOAD FOR
Fable Safe Prompt
Skill for Claude Code
⬇ Download Fable Safe Prompt skill v1.0.0 (.zip)Unzip it, then paste one line into Claude Code (below).
Install it
Fable Foreman v0.7.0 keeps Fable 5.1 or Opus 5.5 planning, routing, and verifying. Fable Safe Prompt rewrites a task so Fable 5 stays Fable. Codex CLI is optional extra workers for the typing.
Codex is optional — skip this block if you don't have a Codex account.
Use it
Run /fable-safe-prompt in an Opus 4.8 session. Paste the result into Fable 5, then put Foreman on the build.
What it does
Fable 5.1 or Opus 5.5 leads Foreman v0.7.0 lets the lead be Fable 5.1 or Opus 5.5. The lead plans the work and personally accepts the result.
Stay on Fable 5 Fable Safe Prompt rewrites a task in clear, neutral language so Fable 5's cybersecurity and biology safety checks do not misread ordinary work and quietly move you onto Opus 4.8 and leave you there.
Optional Codex workers If the Codex CLI is installed, Foreman can hand coding tickets to it. Without it, Foreman routes those tickets to other workers. A tiny test call says whether Codex or Grok can really do work, not just whether you are signed in. The verdicts are LIVE, SIGNED OUT, or BALANCE EXHAUSTED.
Route from one dated table One dated evidence table compares Claude Opus 5.5, Fable 5.1, Sonnet 5.5, and Haiku 5.5, OpenAI GPT-6.1 Sol, GPT-6 Astra, and GPT-6 Luna, and xAI Grok 4.7. The lead always uses the newest release in each model line your accounts can reach (the expensive Astra and Fable lines only when you ask). One command on your machine shows which model gets each job, based on what is actually installed, signed in, and approved on your computer. Clear-cut jobs get a fixed default. Mechanical edits go to GPT-6 Luna. When two good options are close, the lead asks you once, recommends one, and remembers your answer for the rest of the session. An unattended run uses a sensible default and tells you what it chose.
Expensive helpers only when you ask Codex Astra (any version) and Claude Fable are never picked automatically. The lead uses one only if you ask for it and then confirm again once it names the model and its cost. If Fable 5.1 itself is leading your session, that's fine; the rule is about the helpers it hires.
Use Jev only with your own key Jev is a decision model, not a chatbot. With your own TypeSafe or OpenRouter key, the lead uses it for cheap, narrow sorting jobs: merging duplicate review findings, and checking whether cited evidence really supports a finding. That costs a tiny fraction of a normal AI call. Jev only reorders what the lead reviews. It never accepts work or makes security decisions. It asks for the newest Jev release (Jev 1.14 today), starts in a trial mode, and the skill works the same without it.
Check the work before it is done The lead runs the real tests. An independent reviewer, always Claude Opus, checks behavior-changing work. When Claude built the change, a different AI family reviews it when Codex or Grok is available. The lead personally accepts the result.
Checked on real Claude Code runs A behavioral test suite runs real Claude Code leads through 13 scenarios, including picking the newest Codex Sol, a newer Grok that fails its test call, a request for Fable, and a set of Jev situations. It grades their actual tool calls against hidden answer keys. Real Claude Opus 5.5 leads passed every scenario on their latest run, and no run sent work to Astra or Fable or used an outdated model. Those were single runs, not a statistical rate.
Run the Safe Prompt in Opus 4.8 — Fable can trip on the list of flagged words inside the skill — and only for legitimate work. Foreman only uses Codex through the Codex CLI, and it asks before spending your OpenAI quota.
Still deciding? These are the usual questions.
Don't sleep on what's next. Free tools, prompts, and skills for Claude Code and OpenClaw — sent when they drop. No spam, unsubscribe anytime.
Your best model is worth more than grunt work. Put it in charge.
Don't sleep on AI.

