Files
pn-new-crm/src/lib/services/expense-ocr.service.ts
Matt Ciaccio e77d55ac50 feat(insights): Phase B schema + service skeletons
PR1 of Phase B per docs/superpowers/specs/2026-04-28-phase-b-insights-alerts-design.md.
Lays the foundation that PRs 2-10 will fill in with behaviour.

Schema (migration 0014):
- alerts table with rule-engine fields (rule_id, severity, link,
  entity_type/id, fingerprint, fired/dismissed/acknowledged/resolved
  timestamps, jsonb metadata). Partial-unique fingerprint index keeps
  one open row per (port, rule, entity); separate indexes power
  severity-filtered and time-ordered queries.
- analytics_snapshots (port_id, metric_id) -> jsonb cache + computedAt
  for the 15-min recurring refresh.
- expenses: duplicate_of self-FK, dedup_scanned_at, ocr_status/raw/
  confidence; partial index on (port, vendor, amount, date) where
  duplicate_of IS NULL drives the dedup heuristic.
- audit_logs.search_text: GENERATED ALWAYS tsvector over
  action+entity_type+entity_id+user_id, GIN-indexed (drizzle can't
  model GENERATED ALWAYS in TS yet, so the migration appends manual
  ALTER + the GIN index).

Service skeletons in src/lib/services/:
- alerts.service.ts: fingerprintFor, reconcileAlertsForPort (upsert +
  auto-resolve), dismiss, acknowledge, listAlertsForPort.
- alert-rules.ts: RULE_REGISTRY of 10 rule evaluators (currently no-op);
  PR2 fills in the bodies.
- analytics.service.ts: readSnapshot/writeSnapshot with 15-min TTL +
  no-op compute* stubs for the four chart series; PR3 fills behavior.
- expense-dedup.service.ts: scanForDuplicates + markBestDuplicate
  using the partial dedup index. PR8 wires the BullMQ trigger.
- expense-ocr.service.ts: OcrResult/OcrLineItem types + ocrReceipt
  stub. PR9 wires Claude Vision (Haiku 4.5 + ephemeral system-prompt
  cache).
- audit-search.service.ts: tsvector @@ plainto_tsquery + cursor
  pagination on (createdAt, id). PR10 wires the admin UI.

tsc clean, lint clean, vitest 675/675 (one unrelated AES random-output
flake passes solo).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-28 14:43:01 +02:00

48 lines
1.2 KiB
TypeScript

/**
* Claude Vision-driven OCR for expense receipts. PR1 stub: types and the
* service contract. The actual API call wires up in PR9 with prompt
* caching on the system text and Haiku 4.5 by default.
*/
export interface OcrLineItem {
description: string;
quantity?: number;
unitPrice?: number;
amount: number;
}
export interface OcrResult {
vendor: string | null;
amount: number | null;
currency: string | null;
/** ISO date YYYY-MM-DD. */
date: string | null;
lineItems: OcrLineItem[];
/** 0..1; below 0.6 surfaces "verify mode" UI. */
confidence: number;
}
export interface OcrContext {
fileId: string;
fileUrl: string;
/** Optional MIME hint; the service still detects from bytes. */
mimeType?: string;
}
/** Cost ceiling per call (Haiku 4.5 + cached system prompt). PR9 enforces. */
export const OCR_MAX_TOKENS = 1024;
export const OCR_LOW_CONFIDENCE_THRESHOLD = 0.6;
/** Stub — returns "pending" shape so callers can wire UI in PR1 without
* Anthropic credentials. */
export async function ocrReceipt(_ctx: OcrContext): Promise<OcrResult> {
return {
vendor: null,
amount: null,
currency: null,
date: null,
lineItems: [],
confidence: 0,
};
}