changes

fusion_accounting/services/prompts/domain_prompts.py

@@ -18,6 +18,54 @@ You are helping with Canadian HST/GST tax management.
 - Net HST = Collected - ITCs. Positive means owing to CRA.
 - Quarterly filing periods. Check for missing tax on invoices/bills.
 - All vendor bills should have ITCs unless explicitly exempt.
+- HST Purchase tax ID is 20 (13%). No Tax Purchase ID is 32 (0%).
+
+HST FILING WORKFLOW (4 phases — follow this order):
+
+PHASE 1 — REPORTS: Run all at once:
+  calculate_hst_balance, get_tax_report, find_missing_itc_bills,
+  find_missing_tax_invoices, audit_tax_compliance.
+  Present summary with HST position (owing vs refund).
+
+PHASE 2 — BANK SWEEP: Check ALL bank accounts for unreconciled expenses:
+  Call get_unreconciled_bank_lines for each bank journal (RBC Chequing 9595=53,
+  Current Account Scotia=50, Scotiabank Passport Visa 8046=51, RBC Visa X 6752=28).
+  Present ALL unreconciled expense lines (negative amounts) as a fusion-table
+  with your recommendation per row.
+
+PHASE 3 — PER-LINE PROCESSING: For each flagged expense line:
+  0. FIRST: check_recurring_pattern(line_id=X) — if match found, follow action_note
+     instructions EXACTLY (account, HST, partner, reconcile model). No user input needed
+     for recurring payments. If a reconcile_model_id is returned, use apply_reconcile_model.
+  1. get_bank_line_details — check if a vendor bill already exists for same amount/date
+  2. find_similar_bank_lines — check history AND vendor_tax_pattern for coding/tax pattern
+  3. CRITICAL: Check vendor_tax_pattern.is_po_vendor flag:
+     - If is_po_vendor=true: This vendor's bills come from Purchase Orders. Do NOT create
+       a new bill. Instead, use get_unpaid_bills to find the existing bill and propose
+       match_bank_line_to_payments to match the bank payment to that bill.
+     - If is_po_vendor=false: Proceed with bill creation workflow below.
+  4. If bill already exists → propose match_bank_line_to_payments
+  5. If no bill but history match → propose create_vendor_bill with same coding pattern
+  6. If no bill and no history → ask user: "Does this expense include HST?"
+  7. search_partners — find the vendor by keyword from the bank description
+  8. Once confirmed → create_vendor_bill + register_bill_payment (Tier 3, needs approval)
+  9. Alternative: user can choose "Direct GL" → create_expense_entry (Tier 3)
+  For expenses that obviously have no HST (bank fees, interest charges, insurance),
+  proactively recommend "No HST" and explain why.
+
+PO-TRACKED VENDORS (do NOT create bills for these — bills come from Purchase Orders):
+  When find_similar_bank_lines returns is_po_vendor=true or the vendor_tax_pattern
+  note starts with "PO-TRACKED VENDOR", the bill already exists or will be created
+  from a PO. Your job is ONLY to find the existing unpaid bill and match the bank
+  payment to it. If no unpaid bill exists, flag it for the user: "This is a PO vendor
+  but no matching bill was found — the PO may not have been billed yet."
+
+PHASE 4 — VERIFICATION: Re-run calculate_hst_balance and get_tax_report
+  to show the updated HST position after all expenses are recorded.
+
+BANK JOURNAL IDS: RBC Chequing 9595=53, Current Account Scotia=50,
+  Scotiabank Passport Visa 8046=51, RBC Visa X 6752=28.
+MISC JOURNAL: ID=3 (for direct GL expense entries).
 """,
 
     'accounts_receivable': """
@@ -105,5 +153,36 @@ PAYROLL MANAGEMENT CONTEXT:
 }
 
 
+# A3/A5: Aliases so common domain variations still match a prompt
+DOMAIN_ALIASES = {
+    'bank': 'bank_reconciliation',
+    'bank_recon': 'bank_reconciliation',
+    'hst': 'hst_management',
+    'gst': 'hst_management',
+    'tax': 'hst_management',
+    'ar': 'accounts_receivable',
+    'receivable': 'accounts_receivable',
+    'ap': 'accounts_payable',
+    'payable': 'accounts_payable',
+    'journal': 'journal_review',
+    'close': 'month_end',
+    'month_end_close': 'month_end',
+    'payroll': 'payroll_management',
+    'payroll_verify': 'payroll_verification',
+    'stock': 'inventory',
+    'cogs': 'inventory',
+    'report': 'reporting',
+    'reports': 'reporting',
+    'financial': 'reporting',
+}
+
+
 def get_domain_prompt(domain):
-    return DOMAIN_PROMPTS.get(domain, '')
+    if not domain:
+        return ''
+    # Try exact match first, then aliases
+    prompt = DOMAIN_PROMPTS.get(domain, '')
+    if not prompt:
+        resolved = DOMAIN_ALIASES.get(domain, domain)
+        prompt = DOMAIN_PROMPTS.get(resolved, '')
+    return prompt

fusion_accounting/services/prompts/system_prompt.py

@@ -31,12 +31,56 @@ RESPONSE FORMATTING:
 - Use rich Markdown formatting in your responses. The chat renders Markdown as HTML.
 - Use **bold** for account names, amounts, and key terms.
 - Use ## and ### headers to organize sections in longer responses.
-- Use Markdown tables for tabular data (| col1 | col2 | format).
 - Use bullet lists (- item) for findings, issues, and action items.
 - Use numbered lists (1. item) for sequential steps or ranked items.
 - Use `code` for account codes, reference numbers, and technical IDs.
 - Use --- horizontal rules to separate sections in long reports.
 
+INTERACTIVE TABLES (fusion-table) — MANDATORY FOR ACTIONABLE DATA:
+IMPORTANT: When a tool returns a list of records that the user could act on, you MUST use
+a ```fusion-table block instead of a Markdown table. This is REQUIRED — never use plain
+Markdown tables for actionable data. The fusion-table renders an interactive widget with
+checkboxes, your AI recommendations per row, user input fields, and bulk action buttons.
+
+YOU MUST USE fusion-table FOR: missing ITCs/tax (find_missing_itc_bills, find_missing_tax_invoices),
+duplicate entries (find_duplicate_bills, find_duplicate_entries), overdue invoices (get_overdue_invoices),
+unreconciled lines (get_unreconciled_bank_lines, get_unreconciled_receipts, get_unmatched_payments,
+find_unreconciled_suspense), draft entries (find_draft_entries), wrong balances
+(find_wrong_direction_balances), sequence gaps (find_sequence_gaps), wrong accounts
+(find_wrong_account_entries), unpaid bills (get_unpaid_bills), and any other list where
+the user needs to review, dismiss, flag, or create rules for individual rows.
+
+USE REGULAR MARKDOWN TABLES ONLY FOR: P&L (get_profit_loss), balance sheet (get_balance_sheet),
+trial balance (get_trial_balance), cash flow (get_cash_flow), period summaries, tax reports,
+and any purely informational/read-only data where there is nothing to act on per row.
+
+Format: wrap a JSON object in a ```fusion-table fenced code block:
+
+```fusion-table
+{
+  "mode": "interactive",
+  "title": "Descriptive Title",
+  "columns": ["Col1", "Col2", "Col3"],
+  "rows": [
+    {"id": 123, "cells": ["val1", "val2", "val3"], "recommendation": {"action": "dismiss", "reason": "Brief explanation"}},
+    {"id": 456, "cells": ["val1", "val2", "val3"], "recommendation": {"action": "flag", "reason": "Brief explanation"}}
+  ],
+  "actions": ["dismiss", "flag", "create_rule"],
+  "source_tool": "tool_name_that_produced_this"
+}
+```
+
+- "mode": "interactive" (actionable) or "readonly" (informational but structured)
+- "id": the Odoo record ID (account.move id, account.bank.statement.line id, etc.)
+- "recommendation.action": one of "dismiss", "flag", "create_rule"
+- "recommendation.reason": short explanation of why you recommend this action
+- "actions": which bulk action buttons to show
+- "source_tool": the tool name that produced the data
+- You MUST provide a recommendation for each row when using interactive mode.
+- Format monetary amounts as "$X,XXX.XX" in cells.
+- Always include the record ID so actions can target the correct Odoo record.
+- Add a brief text summary before or after the fusion-table block for context.
+
 LINKING TO ODOO RECORDS:
 - When referencing specific records, include clickable Odoo links.
 - Journal entries: [INV/2026/00123](/odoo/accounting/123) where 123 is the move ID.
@@ -60,12 +104,14 @@ def _build_rules_section(rules):
     for rule in rules:
         priority = 'ADMIN' if rule.created_by == 'admin' else 'AI'
         tier = 'auto' if rule.approval_tier == 'auto' else 'needs-approval'
+        conf_str = f', confidence={rule.confidence_score:.0%}, uses={rule.total_uses}' if rule.total_uses > 0 else ''
         lines.append(
-            f'- [{priority}/{tier}] {rule.name} ({rule.rule_type}): '
+            f'- [{priority}/{tier}{conf_str}] {rule.name} ({rule.rule_type}): '
             f'{rule.description or rule.match_logic or "No description"}'
         )
         if rule.match_logic:
-            lines.append(f'  Match logic: {rule.match_logic}')
+            logic_text = rule.match_logic[:500]  # Prevent prompt bloat
+            lines.append(f'  Match logic: {logic_text}')
     return '\n'.join(lines)
 
 
@@ -73,7 +119,9 @@ def _build_history_section(history):
     if not history:
         return ''
     lines = ['RECENT MATCH HISTORY (learn from these patterns):']
-    for h in history[:50]:
+    # A4: Don't hard-cap at 50 — the caller (_load_match_history) already
+    # respects the history_in_prompt config setting
+    for h in history:
         status = h.decision
         reason = ''
         if h.rejection_reason: