Turns the selector's raw confidence into a score that means what it says.
- lib/agent/categorize/calibration.ts: the engine. Isotonic regression
(pool-adjacent-violators, distribution-free + monotonic) over
(confidence, was_correct) samples → a calibrator; plus reliabilityByBucket,
ECE, and bandFor(). bandFor NEVER returns 'auto' without a fitted calibrator
(no silent booking on an unproven score) and never auto-books above an amount
cap. 12 engine tests (overconfidence pulled down, underconfidence lifted,
monotonicity, ECE, band gating).
- Measurement loop: migration categorize_calibration_samples (append-only,
company-scoped RLS, confidence CHECK [0,1]) + POST /api/agent/categorize/
outcome logging one sample (proposed vs actually booked) fire-and-forget from
QuickReviewDialog on a successful book (sandbox skipped). AiCategorizeProposal
surfaces the proposal metadata via onProposal.
- scripts/fit-categorize-calibration.ts (read-only): prints the reliability
diagram + ECE + fitted calibrator once data has accumulated.
Fitting needs a few hundred real outcomes, so nothing calibrates today — the
loop starts collecting, and "säker" stays uncalibrated (no auto-book) until the
data proves it. 131 unit tests green; RLS covered by a pg-real test.
Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Wires the auto-booking cascade to the UI. When the quick-review dialog opens
for a transaction, AiCategorizeProposal fetches POST /api/agent/categorize
(Tier 1 deterministic candidates → Tier 2 model selector, provider-agnostic)
and:
- pre-fills the dialog's account + VAT from the model's pick (through
handleAccountChange, so class-2 VAT clearing is preserved),
- shows a confidence band (säker / trolig / välj konto), a short "Varför", and
the candidate alternatives (click one to re-apply),
- falls back SILENTLY to the deterministic defaults on error, and shows a soft
note when no AI backend is configured — the dialog always works without AI.
Surfaced inside the existing dialog (not a new card) so it reuses the proven,
balanced commit path. No silent auto-posting: "säker" is pre-filled and booked
with the dialog's existing one-tap Bokför; hands-off auto-book waits for
calibration (avoids any storno-on-undo).
lint + guards clean; 94 transaction/categorize tests green (no regressions).
Strings inline Swedish for now (lift to messages/{sv,en} pre-merge). UI PR:
awaiting founder visual sign-off before merge.
Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
