Image: GitHub
Why it mattersA team paying for Codex on either the Pro or the Plus plan now has a shared, documented pattern for splitting the expensive planning model from the cheaper execution model, and a script that reports what each thread actually cost.
donvito/codex-astra-luna-orchestrator, a Codex configuration package that installs a hierarchical setup with GPT-6 Astra as the root orchestrator and GPT-5.6 Luna as the default subagent, reached 971 stars on GitHub in the six days since it was created on 5 September, per the repository's public star count and creation date on the GitHub API. The author describes it as "a configurable Codex setup where GPT-6 Astra is the root/orchestrator and reviewer, while GPT-5.6 Luna is the default and pinned model for execution subagents."
Two plans, one pattern
The package includes two profiles, matched to the two paid Codex plans. On the Pro plan, Astra is pinned to the root at medium reasoning effort and Luna takes the explorer, worker, tester and researcher subagent roles at maximum effort. On the Plus plan, Luna runs the root at maximum effort as well, because the author's note reads: "Plus users: the root thread is the largest line item, so running it on Luna saves the most." A separate reviewer role runs at low effort in both plans, on Astra, and acts as a final independent check on subagent output.
The pattern rests on a claim the author sets out directly in the README: "orchestration is not free: the root stays in the loop for the whole task and every subagent carries its own context." That is why the cheap model does the worker tasks and the expensive one is used only where it decides what happens, or in the reviewer role where it looks at what did happen.
The token accounting is the part worth reading
Under scripts/token_usage.py the repository includes a Python analysis script that reads the rollout logs Codex already writes under ~/.codex/sessions and reports usage per thread, per role, and per model, plus the change in the account's 5-hour and 7-day rate limit windows. That data shows whether the split actually saves money on a given type of task: without it, the choice between Pro and Plus rests on which model seems smarter, and ignores which pattern spends fewer tokens for the same outcome.
What it does and does not do
The package installs skills, agent role profiles as TOML files, and a AGENTS.md orchestration guide, so it is a set of configuration files and skills, and it adds no new agent runtime. It runs on top of Codex, which does the work of routing tool calls, holding sessions and issuing prompts. That is deliberate: the codex-astra-luna-orchestrator repo aims to be the shared, versioned answer to "how should I set up Codex with these two models", and leaves the rest to Codex.
For a team already on Codex, that is the practical value. Two engineers reading the same short README can set up the same orchestrator/worker split and look at the same numbers to decide whether the reviewer step or the max-effort subagents were worth their token cost. That costs less than each team writing the split from scratch.
Source
The repository is donvito/codex-astra-luna-orchestrator on GitHub. Star count and creation date verified against the public repository API at the time this was published.
This item was written by an AI system from the linked source. Reveneau is responsible for what it publishes.
Get AI News in your inbox
New developer tools, model and agent releases, and how teams are actually using them to release software. Short, and only when there is something worth reading.