Add the decision-model eval runner History row to the os-me status leaf #1300

New issue

Open

opened 2026-10-23 23:02:45 +00:00 by jared · 0 comments

jared commented

2026-10-23 23:02:45 +00:00

Owner

Problem

decision-model ticket #1 (eval runner, children #2-#6) closes with the item "update the status leaf". The only status leaf that tracks the decision-model gem is docs/implementation-status/os-me.md. Its last History row is the 2026-10-23 S8 gem-move row. No row records the eval runner.

What shipped (decision-model main, 2026-10-23)

#2 c628baa: decision-eval <dir> CLI, Kappa, ConfusionMatrix, report.json and report.md.
#3 c04b4d6: outside adapters loaded from config by require path and class.
#4 6f0feb8: tune, holdout and frozen splits; --frozen; frozen-reads.jsonl.
#5 7efe78c: free-text judge mode; agreement and kappa per rule.
#6 ce2b606: context limit per adapter (fail loudly); CriteriaRules pre-check at suite load.
d2e04bd: README and CHANGELOG for the additions above.

Work

Add one History row for the eval runner to docs/implementation-status/os-me.md, or name the leaf that should hold it.

Discoverer: /home/jared/dev/ruby-gems/decision-model, session b6d48eed-a394-4b5d-ac38-c8bee60b0902, 2026-10-23. Ticket #1 close-out names a status leaf that lives in cc-os.

## Problem decision-model ticket #1 (eval runner, children #2-#6) closes with the item "update the status leaf". The only status leaf that tracks the decision-model gem is `docs/implementation-status/os-me.md`. Its last History row is the 2026-10-23 S8 gem-move row. No row records the eval runner. ## What shipped (decision-model main, 2026-10-23) - #2 c628baa: `decision-eval <dir>` CLI, Kappa, ConfusionMatrix, report.json and report.md. - #3 c04b4d6: outside adapters loaded from config by require path and class. - #4 6f0feb8: tune, holdout and frozen splits; `--frozen`; `frozen-reads.jsonl`. - #5 7efe78c: free-text judge mode; agreement and kappa per rule. - #6 ce2b606: context limit per adapter (fail loudly); CriteriaRules pre-check at suite load. - d2e04bd: README and CHANGELOG for the additions above. ## Work Add one History row for the eval runner to `docs/implementation-status/os-me.md`, or name the leaf that should hold it. -------- **Discoverer:** /home/jared/dev/ruby-gems/decision-model, session b6d48eed-a394-4b5d-ac38-c8bee60b0902, 2026-10-23. Ticket #1 close-out names a status leaf that lives in cc-os.