Frame ML Problem

Write ## Modeling decisions in journal/JOURNAL.md. The table is the contract. This skill does not declare a learner and does not evaluate one.

Human-facing prose

Details: setup-workspace references/human_facing_prose.md. Journal cells describe this dataset. Do not name the skills framework, the CLI, or a splitter class in the table. Questions use data-science language — not skill ids, G-* names, or the wrapper CLI.

Procedure

1. Run python -m skore_skills status. If has_journal is false, send the user to setup and stop. If data_analysis is missing and explore-ml-data is installed, AskUserQuestion: explore first (default) or continue from facts the user stated. Explore loads explore-ml-data and stops. Do not invent dataset facts. Do not ask this again once data_analysis is present or skipped.
2. Run python -m skore_skills frame show. When the user is changing a locked constraint and named one cell, add --revise. When they are changing a constraint and did not name a cell, ask which filled decision to change (skip n/a) and stop. Do not --revise, do not frame clear, and do not edit the journal on that turn. JSON action is authoritative. Do not invent a menu. If the command is missing or exits without JSON, read references/fallback.md and follow it. Do not open another reference. Do not guess candidates.
3. stop — say the JSON reason and stop.
4. ask / uncovered — read references/fallback.md only. Write Prediction goal uncovered and the prose cells it names. Set Status to draft. Stop this turn.
5. ask / missing_keys — read each distinct reference in questions once before asking. Do not open any other file under references/. Ask every key in missing in one message. For a question that has candidates, those are the options. When candidates is absent, ask for the value the reference describes. Draw on three sources, and only what they actually say: the EDA report, free-form text that came with the data if any is present (notes, a dictionary, or a README beside the raw files), and facts the user stated. If none of that text is present, do not invent it. When one of them already states the fact, quote it in the question. Write every Value cell the user answered in this turn. Do not rename Variable cells. Do not stop after the first cell. When the deployment makes other rows inapplicable, set those cells to n/a in the same edit. Horizon, gap, and time role are n/a unless deployment is time. Generalize-to is n/a unless deployment is groups. A fold count of 1 is one train/test split. Once any decision cell is filled and Status is not locked, set Status to draft. Stop this turn.
6. When the user named one cell and Status is draft, do not use the lock menu as the change. Run python -m skore_skills frame clear --cell <key> for that cell and stop. Do not write the new value. Do not name any other cell as cleared. The command's JSON blanked list is the record. Status stays draft. The next frame show asks only keys that are still empty or invalid.
7. ask / confirm_lock or ask / revise — quote JSON context inline, then offer JSON choices only and stop. The user sentence that opened this screen is not a choice. Do not set Status to locked in that same turn.
   • lock on a later turn sets Status to locked.
   • modify on a revise, when the user named one cell: run python -m skore_skills frame clear --cell <key> and stop. Do not write the new value. Do not blank any other cell by hand. The next frame show asks only keys that are still empty or invalid. modify with no named cell writes nothing and does not frame clear: ask which filled decision to change and stop.
   • keep leaves the locked table unchanged.
   • stop writes nothing further.
8. proceed — the table is locked. If translation is null, say that this lock has no splitter translation. Do not load build-ml-pipeline and do not return to model-ml-pipeline. Stop. If model-ml-pipeline dispatched this turn, return to that coordinator and stop. Do not start build, write a design note, or run the git close from here. Otherwise run python -m skore_skills git end-turn --stage implement. If JSON action is invoke, load persist-ml-git only if status.skills.persist-ml-git is true and stop. Otherwise load triage-ml-task only if that skill is installed.

Stop conditions

• Do not write Python, a pipeline, a test, or a design note.
• Do not put a class name or a constructor argument in the journal. TimeSeriesSplit, KFold, GroupKFold, and gap= stay out of the table.
• Do not re-ask a key that is absent from missing.
• Do not add an option that is absent from candidates.
• Do not open a reference the JSON did not name, except references/fallback.md when the command is missing.
• A locked table changes only through frame show --revise, then the same fill and confirm gates. keep does not edit it.
• On modify, frame clear is the only journal edit, and only for the cell the user named. Do not rewrite experiments/, audit/, or a report in this skill.
• After a cell is blanked, do not run an existing experiment script. Say that it still uses the previous splitter and metric. The next build or evaluate rewrites it after the table is locked again.