No study yet: this module has no plate position.
What it does for you: It is the one place that says which parts of the Factory may use an AI model, which model, how hard it should think, and why. Most parts never use one: recording, checking, running, isolating and showing work must give the same answer every time, and a model's opinion is not evidence. Where a model can help (proposing edits, and later reviewing them or drafting summaries and plans), the task names exactly one model and effort level, the check that must accept its answer, and what happens when there is no answer. To save cost it lowers the effort level first; a cheaper model is used only once a measured comparison shows it does the same job for less.
How it links: Covers every module in the Factory, including planned ones such as the nine in the proposed discovery and post-development design, and a check fails if a new part arrives without saying whether it uses a model. It agrees with the Producers module: the same size limits, the same model for proposals. Nothing uses it yet: the Product gate does not consult it, and the planned delivery flow will ask it which model a task may run before anything is sent.
Honest limits:
- Nothing is qualified, so nothing runs. Claude Code is waiting for a fresh check under its current rules. OpenAI models are now meant to be reached through the Codex command line on your ChatGPT sign-in, never a Platform key; that route is only named so far, not built or checked, and the older Platform-key connection is kept but no longer used. The Codex command line has no dedicated setting for how hard the model thinks (only a general configuration override, which the planned set-up does not use), so the design's three checks by an OpenAI model (auditing the sorting, challenging proposals and reviewing release drafts) are listed as blocked: named, with the intended model and effort, but always refused until that route has its own check.
- The planned discovery and post-development parts would use a smaller, cheaper Claude model for sorting, searching and drafting. That is a guess until it is measured against the larger model on the same work, and the Factory refuses to run each one until its own measurement is recorded. The weekly web search would use a separate Claude set-up with a search tool, which stays unavailable until it has its own safety check. Each choice of model follows a default that you have not yet signed.
- Only proposing edits has a finished question-and-answer format. Review, summaries, briefs, planning and every planned discovery and post-development task each need their own before they can run.
- The choices of model and effort are reasoned guesses, each with a measure that would retire it; none has been measured yet. Model prices and limits are a dated copy, not live.
- It decides nothing by itself and grants nothing: a model's answer is only ever a proposal that other checks accept or reject.
Status: Accepted: independent code review (Astra high) passed on 29 September 2026 in round 2. Three changes for the proposed discovery and post-development design each passed their own independent check since: increment 0 adds planned entries, a discovery allowance and the Codex route (round 2); increment 1 moves the Lifecycle schedule's entry beside its new code, unchanged (round 4); and increment 2 records that the page reader, Discovery sources, has been built and uses no model (round 5). Joining them with the main line moves the planned entries out of the README's table into the agent reference, so the table fits its 25 lines; that change is checked with the join. Joining Service access, whose entry uses no model, adds one row, which fills the table's 25 lines; that join awaits its own independent check. No model is qualified yet, so every task refuses today. This is local development only: not a release, a Verdict or a customer benefit, and all 27 epics remain open.