Commit graph

5 commits

Author SHA1 Message Date
13f6c28dc0
fix: findings from the Cowork test, released as v0.1.1
The operator tested v0.1.0 in Cowork on 2026-11-11. README now says what
was measured there and what was not.

- miljo: no longer claims that kontroll compares "with tolerance, not bit
  for bit" when numpy or scipy differ from the note's versions. kontroll
  always compares byte for byte first, and numpy 2.5.3 gave bit for bit.
- kontroll --alt: show every differing line. The cut-off line now names
  the option instead of only counting the lines left out.
- skill simuler: the question before a run must be answerable with yes or
  no. In the test it became a choice between variants and the run stalled.
- skill forbedre: offer the working copy as a zip when the user cannot
  open the folder it was made in, as in Cowork.

263 tests pass; full kontroll: bit for bit with the note.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-11 20:24:58 +01:00
19858bbda9
docs: bring skills and docs in line with the fixed code
Skills ask before using a working copy found in the user's folder and state
its check level first. The drift of the queue models before the shock is
stated in the caveats and the README. README shows how to run the check and
the tests. The drift line names the month a gradual drive begins.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-11 15:46:39 +01:00
776325c680
fix: findings from the independent review
Scenario runner: stops when overridden base parameters leave the start out
of balance (new field startlikevekt: krev, kalibrer, ubalanse); stops on
non-finite results; says that models with a court queue drift slowly even
without a shock and prints the drift before the shock; queue in months of
inflow; never deletes files it did not write; stricter limits.
Check: anchored on the note's own checksums for core and reference files,
runs the scenario files too, exact comparison for the diagnosis, honest
wording when parts were skipped or an extension is broken.
Launcher: unknown field names in input are rejected with a suggestion;
unwritable folders give a message, not a traceback.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-11 15:27:14 +01:00
03151cfb3b
feat(scenario): scenario runner, extensions and working copy
scenario.py runs all five simulators of the core from a JSON scenario file,
with sweeps, and reproduces 20 of the note's data files (14 byte for byte,
6 column-wise). Extensions live outside the core and must reproduce it
exactly in a limit case; paadrag_tidsserie is the worked example.
arbeidskopi.py gives the user a copy of the model they can change.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-11 14:17:32 +01:00
464ad38b49
feat: model core, launcher, reproduction check and the first two skills
The core in modell/kjerne/ is an unmodified copy of sim.py, diagnose.py and
forlop.py from den-dynamiske-rettsstaten @2badc25, pinned by sha256.
kjor.py is the single entry point; kontroll.py reproduces the note's 22 data
files and printed results. Skills: diagnose and forlop (Norwegian).

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-11 13:53:32 +01:00