Claude Fable handles planning and review; GPT-5.5 Codex handles implementation and research. Two Claude Code skills wire that split into a repo-centered loop: specs and gates are written first, Codex works in fresh contexts, and Fable reviews the evidence before anything is integrated. It runs on the subscriptions you already have — no API keys required by default.

git clone https://github.com/DanMcInerney/architect-loop
cd architect-loop && ./install.sh        # Windows: .\install.ps1
npm i -g @openai/codex@latest            # the builder (Codex CLI >= 0.133)

./install.sh --project installs to the current repo only instead of globally. You need Claude Code on any paid plan and the Codex CLI signed into a ChatGPT plan.

/architect                                      # the build loop
/architect-research <what you're considering>   # the research loop

/architect runs one work block: judge the last run, spec the next slice, dispatch builders. /architect-research is for when you're still deciding what to build — its cited report feeds the build loop's PRD.

One short Fable session per work block — judgment only, it never writes code:

• Spec + gates first. Fable specs a one-PR slice, splits it into 1–4 lanes whose file sets are checked for overlap, and commits the acceptance gates to docs/gates/ before any builder starts. Gates are read-only; a builder edit to a gate file fails the slice automatically.
• Parallel isolated builders. One fresh codex exec (xhigh) per lane, each in its own git worktree. Builders must argue with the spec before building (silent compliance = defect), build only their declared files, and report raw results — they do not have commit access in the sandbox.
• Fable judges and integrates. It runs the gate commands itself (builder claims are hearsay), reads the diff against the spec's intent (passing tests ≠ mergeable work), then commits and merges passing lanes. Judgment happens in a fresh session because the cited evidence favors fresh-context review.
• The repo is the only memory. docs/HANDOFF.md (a short table of contents, pruned every session), docs/gates/, docs/lanes/, git history. Not in the repo = didn't happen.
• Supervision built in. Liveness checks on dispatched runs, stall triage (diagnose the child process tree, kill the narrowest thing), explicit timeouts on every long command.

Scout-first, like the production deep-research systems — no fixed lane taxonomy:

• A cheap Codex scout maps the topic (~10 searches): canonical terminology, the load-bearing systems and papers, the named people, the topic's natural fault lines. Skipped for comparisons and fact-finds.
• Fable designs 3–6 topic-specific lanes from the scout's map, drawing per-source-class tactics from a library (academic citation snowballing, dependents-not-stars repo evidence, emerging-vs-hype gating, production pattern mining, expert tracking) — checked for overlap and gaps before dispatch.
• Parallel Codex researchers run under hard budgets: search caps, ≤5 subjects per lane, saturation stop, strict findings discipline (URL + date
  • quote + confidence tag; NOT FOUND beats inference; no recommendations). Expert opinion runs as a second wave, roster-seeded by the first.
• Fable verifies and writes. ≥2 independent sources per load-bearing claim, adversarial falsification searches, citations only from URLs actually fetched — then one author writes one decision-oriented report. Gathering parallelizes; synthesis never does.

Each design choice is source-backed (full citations in DESIGN.md):

• Weak planners hurt more than weak executors — so the architect model does the design, and builders get explicit specs.
• Manager + worktree-isolated workers is a well-supported topology for shared-artifact software work; naive shared-file coordination collapses throughput.
• Frozen external gates beat trusting the agent — but agents game visible tests and their passing PRs are frequently unmergeable, so the architect also reads the diff.
• Memory files rot — so the handoff stays a short map, and detail lives in linked gate/lane files.
• The surveyed production deep-research systems use planner-designed decomposition rather than fixed lanes — so research lanes are designed per topic, after a scout pass.

Do I need API keys? No. Claude Code runs on your Claude plan; Codex CLI on your ChatGPT plan.

What does a run cost? Builder/researcher runs draw on your ChatGPT plan's 5-hour and weekly quotas; a multi-hour run is a meaningful fraction of a weekly window. Fable's architect sessions are minutes, not hours.

What if a builder wrecks things? Nothing reaches a branch until the architect's tamper, boundary, and gate checks pass — worktrees are discarded and re-dispatched from the freeze commit.

Can I watch a run? Yes — every dispatch prints the builder block, so you can paste it into an interactive codex session with /goal instead.

Why two skills? Research-grade fan-out costs ~15× chat-level tokens — it should be a deliberate act, not a side-effect of the build loop.

The original idea came from this X post by @jumperz about using Fable with Codex subagents. I built architect-loop because I couldn't find an easy way to run that pattern, and because it seemed useful to add a few extra operational best practices on top of what Fable can already do when calling Codex subagents.

MIT