CHINN — Chapter Hierarchy Intelligence Neural Network (retry 1) (retry 2)

{
  "cost": {
    "input_tokens": 2072,
    "estimated_usd": 0.140691,
    "output_tokens": 8965
  },
  "model": "claude-sonnet-4-20250514",
  "status": "success",
  "transport": "anthropic_api",
  "latency_ms": 362099,
  "response_text": "# CHINN Implementation\n\nI'll implement CHINN as a Layer 2 Trade Operations INN that routes classification requests to chapter-specialized logic, mapping to CBP's 10 Centers of Excellence and Expertise.\n\n## 1. Core CHINN Implementation\n\n```python\n# oneworld_trade/agents/chinn.py\n\"\"\"\nCHINN — Chapter Hierarchy Intelligence Neural Network\n\nPhilosophy: \"Classify by domain, not by code. A textile specialist sees fiber content \nwhere a metals specialist sees alloy grade. The chapter determines what matters — \nCHINN routes the question to the expert who knows what to look for.\"\n\nCHINN is a Layer 2 Trade Operations INN that sits between UINN (user intent) and \nthe existing classification INNs (RINN, OINN, FINN), adding chapter-awareness that \nno current INN provides. It maps directly to CBP's 10 Centers of Excellence and \nExpertise (CEEs), organizing trade enforcement by industry domain.\n\nFlow: Product → UINN (intent) → CHINN (chapter router) → Chapter Specialist → RINN/OINN/FINN\n\"\"\"\n\nfrom dataclasses import dataclass, field\nfrom datetime import date, datetime\nfrom enum import Enum\nfrom typing import Dict, List, Optional, Tuple, Union\nimport logging\n\ntry:\n    from ..core.pattern_bus import PatternBus, PatternType\nexcept ImportError:\n    PatternBus = None\n    PatternType = None\n\ntry:\n    from supabase import create_client, Client\n    import os\n    SUPABASE_URL = os.getenv('SUPABASE_URL')\n    SUPABASE_KEY = os.getenv('SUPABASE_ANON_KEY')\n    supabase: Optional[Client] = create_client(SUPABASE_URL, SUPABASE_KEY) if SUPABASE_URL and SUPABASE_KEY else None\nexcept ImportError:\n    supabase = None\n\nlogger = logging.getLogger(__name__)\n\nclass CEEDomain(Enum):\n    \"\"\"CBP's 10 Centers of Excellence and Expertise mapped to HTS chapter ranges\"\"\"\n    PHARMA_HEALTH_CHEMICALS = \"CEE001\"\n    AGRICULTURE_PREPARED = \"CEE002\" \n    AUTOMOTIVE_AEROSPACE = \"CEE003\"\n    APPAREL_FOOTWEAR_TEXTILES = \"CEE004\"\n    BASE_METALS = \"CEE005\"\n    PETROLEUM_GAS_MINERALS = \"CEE006\"\n    ELECTRONICS = \"CEE007\"\n    CONSUMER_MASS_MERCH = \"CEE008\"\n    INDUSTRIAL_MANUFACTURING = \"CEE009\"\n    MACHINERY = \"CEE010\"\n\nclass GRIRule(Enum):\n    \"\"\"General Rules of Interpretation - which dominate by domain\"\"\"\n    GRI1 = \"heading_text_specificity\"\n    GRI2A = \"incomplete_unfinished\"\n    GRI2B = \"mixtures_composites\"\n    GRI3A = \"specific_over_general\"\n    GRI3B = \"material_component\"\n    GRI3C = \"last_numerical_heading\"\n    GRI4 = \"most_similar\"\n    GRI5A = \"containers_together\"\n    GRI5B = \"containers_separate\"\n    GRI6 = \"subheading_rules\"\n\n@dataclass\nclass ChapterSpecialist:\n    \"\"\"Configuration for chapter-specific classification expertise\"\"\"\n    domain: CEEDomain\n    chapter_range: List[str]\n    classification_signals: List[str]\n    primary_gri: List[GRIRule]\n    pga_agencies: List[str]\n    risk_flags: List[str]\n    de_minimis_patterns: Dict[str, Union[str, float]]\n    fta_rules: Dict[str, str]\n    \n    def handles_chapter(self, chapter: str) -> bool:\n        \"\"\"Check if this specialist handles the given HTS chapter\"\"\"\n        return chapter in self.chapter_range\n    \n    def get_classification_priority(self, signals: Dict[str, any]) -> Dict[str, float]:\n        \"\"\"Weight classification signals by domain expertise\"\"\"\n        weights = {}\n        for signal in self.classification_signals:\n            if signal in signals:\n                weights[signal] = 1.0  # Base weight, can be refined per domain\n        return weights\n\n@dataclass\nclass QPCertification:\n    \"\"\"Qualified Person certification tracking\"\"\"\n    qp_name: str\n    certified_chapters: List[str]\n    certification_date: date\n    rulings_count: int = 0\n    accuracy_rate: float = 0.0\n    cee_domain: Optional[CEEDomain] = None\n    \n    def is_certified_for_chapter(self, chapter: str) -> bool:\n        \"\"\"Check if QP is certified for given chapter\"\"\"\n        return chapter in self.certified_chapters\n    \n    def needs_renewal(self) -> bool:\n        \"\"\"Check if certification needs renewal (2 year cycle)\"\"\"\n        from datetime import datetime\n        today = datetime.now().date()\n        years_since = (today - self.certification_date).days / 365\n        return years_since >= 2.0\n\n@dataclass\nclass ChapterRoutingResult:\n    \"\"\"Result of CHINN chapter routing decision\"\"\"\n    product_id: str\n    detected_chapter: str\n    cee_domain: CEEDomain\n    specialist: ChapterSpecialist\n    confidence: float\n    routing_signals: Dict[str, any]\n    escalation_required: bool = False\n    qp_review_needed: bool = False\n    \n@dataclass\nclass ChapterSignals:\n    \"\"\"Chapter-specific classification signals to prioritize\"\"\"\n    chapter: str\n    domain: CEEDomain\n    primary_signals: List[str]\n    secondary_signals: List[str]\n    pga_requirements: List[str]\n    risk_screening: List[str]\n\nclass CHINNCore:\n    \"\"\"Core CHINN routing and specialist management\"\"\"\n    \n    def __init__(self):\n        self.specialists = self._initialize_specialists()\n        self.qp_certifications = self._initialize_qp_certs()\n        self.pattern_bus = PatternBus() if PatternBus else None\n        \n    def _initialize_specialists(self) -> Dict[CEEDomain, ChapterSpecialist]:\n        \"\"\"Initialize chapter specialists based on CBP CEE structure\"\"\"\n        specialists = {\n            CEEDomain.PHARMA_HEALTH_CHEMICALS: ChapterSpecialist(\n                domain=CEEDomain.PHARMA_HEALTH_CHEMICALS,\n                chapter_range=[\"28\", \"29\", \"30\", \"33\", \"34\", \"38\"],\n                classification_signals=[\"active_ingredient\", \"pharmaceutical_grade\", \"chemical_composition\", \"therapeutic_use\"],\n                primary_gri=[GRIRule.GRI1, GRIRule.GRI3B],\n                pga_agencies=[\"FDA\", \"EPA\", \"DEA\"],\n                risk_flags=[\"controlled_substance\", \"fda_approval\", \"chemical_weapons\"],\n                de_minimis_patterns={\"pharma\": 0.0, \"chemicals\": 800.0},\n                fta_rules={\"chemical_transformation\": \"substantial_transformation\"}\n            ),\n            \n            CEEDomain.AGRICULTURE_PREPARED: ChapterSpecialist(\n                domain=CEEDomain.AGRICULTURE_PREPARED,\n                chapter_range=[str(i) for i in range(1, 25)],\n                classification_signals=[\"species\", \"processing_level\", \"preservation_method\", \"food_safety\"],\n                primary_gri=[GRIRule.GRI1, GRIRule.GRI2B],\n                pga_agencies=[\"USDA\", \"FDA\", \"APHIS\", \"FWS\"],\n                risk_flags=[\"lacey_act\", \"cites\", \"bioterrorism\", \"country_of_origin\"],\n                de_minimis_patterns={\"food\": 0.0, \"agricultural\": 800.0},\n                fta_rules={\"agricultural_goods\": \"wholly_obtained\"}\n            ),\n            \n            CEEDomain.AUTOMOTIVE_AEROSPACE: ChapterSpecialist(\n                domain=CEEDomain.AUTOMOTIVE_AEROSPACE,\n                chapter_range=[\"87\", \"88\"],\n                classification_signals=[\"vehicle_type\", \"engine_specs\", \"safety_features\", \"commercial_use\"],\n                primary_gri=[GRIRule.GRI1, GRIRule.GRI2A],\n                pga_agencies=[\"DOT\", \"NHTSA\", \"FAA\", \"EPA\"],\n                risk_flags=[\"safety_standards\", \"emissions\", \"dual_use\"],\n                de_minimis_patterns={\"automotive\": 800.0, \"aerospace\": 0.0},\n                fta_rules={\"automotive\": \"regional_value_content\", \"net_cost_method\": \"62.5%\"}\n            ),\n            \n            CEEDomain.APPAREL_FOOTWEAR_TEXTILES: ChapterSpecialist(\n                domain=CEEDomain.APPAREL_FOOTWEAR_TEXTILES,\n                chapter_range=[str(i) for i in range(50, 68)] + [\"64\", \"65\"],\n                classification_signals=[\"fiber_content\", \"weave_type\", \"garment_construction\", \"gender_age\"],\n                primary_gri=[GRIRule.GRI3B, GRIRule.GRI2B],\n                pga_agencies=[\"CPSC\", \"FTC\", \"CBP\"],\n                risk_flags=[\"UFLPA\", \"cotton_provenance\", \"flammability\", \"country_marking\"],\n                de_minimis_patterns={\"textiles\": 800.0, \"apparel\": 800.0},\n                fta_rules={\"yarn_forward\": \"regional_yarn\", \"fiber_forward\": \"regional_fiber\"}\n            ),\n            \n            CEEDomain.BASE_METALS: ChapterSpecialist(\n                domain=CEEDomain.BASE_METALS,\n                chapter_range=[str(i) for i in range(72, 84)],\n                classification_signals=[\"alloy_grade\", \"metal_composition\", \"form_factor\", \"heat_treatment\"],\n                primary_gri=[GRIRule.GRI1, GRIRule.GRI3B],\n                pga_agencies=[\"DOC\", \"CBP\"],\n                risk_flags=[\"AD_CVD\", \"232_tariffs\", \"CBAM\", \"strategic_materials\"],\n                de_minimis_patterns={\"base_metals\": 800.0},\n                fta_rules={\"substantial_transformation\": \"melting_and_casting\"}\n            ),\n            \n            CEEDomain.PETROLEUM_GAS_MINERALS: ChapterSpecialist(\n                domain=CEEDomain.PETROLEUM_GAS_MINERALS,\n                chapter_range=[\"25\", \"26\", \"27\"],\n                classification_signals=[\"mineral_composition\", \"extraction_method\", \"purity_grade\", \"energy_content\"],\n                primary_gri=[GRIRule.GRI1, GRIRule.GRI2B],\n                pga_agencies=[\"DOE\", \"EPA\", \"BIS\"],\n                risk_flags=[\"strategic_petroleum\", \"export_controls\", \"environmental\"],\n                de_minimis_patterns={\"petroleum\": 0.0, \"minerals\": 800.0},\n                fta_rules={\"wholly_obtained\": \"extracted_in_territory\"}\n            ),\n            \n            CEEDomain.ELECTRONICS: ChapterSpecialist(\n                domain=CEEDomain.ELECTRONICS,\n                chapter_range=[\"84\", \"85\", \"90\"],\n                classification_signals=[\"principal_function\", \"processing_capability\", \"connectivity\", \"end_user\"],\n                primary_gri=[GRIRule.GRI1, GRIRule.GRI3A],\n                pga_agencies=[\"FCC\", \"BIS\", \"CPSC\"],\n                risk_flags=[\"itar\", \"ear\", \"section_301\", \"entity_list\", \"telecommunications\"],\n                de_minimis_patterns={\"electronics\": 800.0, \"computers\": 800.0},\n                fta_rules={\"substantial_transformation\": \"programming_and_testing\"}\n            ),\n            \n            CEEDomain.CONSUMER_MASS_MERCH: ChapterSpecialist(\n                domain=CEEDomain.CONSUMER_MASS_MERCH,\n                chapter_range=[\"39\", \"42\", \"69\", \"94\", \"95\", \"96\"],\n                classification_signals=[\"consumer_use\", \"material\", \"age_group\", \"safety_features\"],\n                primary_gri=[GRIRule.GRI1, GRIRule.GRI3C],\n                pga_agencies=[\"CPSC\", \"FTC\"],\n                risk_flags=[\"toy_safety\", \"flammability\", \"lead_content\"],\n                de_minimis_patterns={\"consumer_goods\": 800.0},\n                fta_rules={\"substantial_transformation\": \"change_in_tariff_classification\"}\n            ),\n            \n            CEEDomain.INDUSTRIAL_MANUFACTURING: ChapterSpecialist(\n                domain=CEEDomain.INDUSTRIAL_MANUFACTURING,\n                chapter_range=[\"39\", \"40\", \"44\", \"68\", \"69\", \"70\"],\n                classification_signals=[\"industrial_use\", \"processing_method\", \"composition\", \"form_factor\"],\n                primary_gri=[GRIRule.GRI1, GRIRule.GRI2A],\n                pga_agencies=[\"EPA\", \"OSHA\"],\n                risk_flags=[\"chemical_safety\", \"environmental\", \"industrial_standards\"],\n                de_minimis_patterns={\"industrial\": 800.0},\n                fta_rules={\"substantial_transformation\": \"manufacturing_process\"}\n            ),\n            \n            CEEDomain.MACHINERY: ChapterSpecialist(\n                domain=CEEDomain.MACHINERY,\n                chapter_range=[\"84\", \"85\", \"90\"],\n                classification_signals=[\"mechanical_function\", \"power_source\", \"automation_level\", \"industry_use\"],\n                primary_gri=[GRIRule.GRI1, GRIRule.GRI3A],\n                pga_agencies=[\"BIS\", \"OSHA\"],\n                risk_flags=[\"dual_use\", \"export_controls\", \"safety_standards\"],\n                de_minimis_patterns={\"machinery\": 800.0},\n                fta_rules={\"substantial_transformation\": \"assembly_and_testing\"}\n            )\n        }\n        return specialists\n    \n    def _initialize_qp_certs(self) -> Dict[str, QPCertification]:\n        \"\"\"Initialize QP certifications - starting with Jack Chang\"\"\"\n        return {\n            \"jack_chang\": QPCertification(\n                qp_name=\"Jack Chang\",\n                certified_chapters=[\"61\", \"62\", \"63\"],  # Apparel chapters in training\n                certification_date=date(2024, 1, 1),\n                rulings_count=0,\n                accuracy_rate=0.0,\n                cee_domain=CEEDomain.APPAREL_FOOTWEAR_TEXTILES\n            )\n        }\n    \n    def route_by_chapter(self, product_description: str, initial_chapter: Optional[str] = None) -> ChapterRoutingResult:\n        \"\"\"Route classification request to appropriate chapter specialist\"\"\"\n        # Determine chapter if not provided\n        if not initial_chapter:\n            initial_chapter = self._detect_chapter(product_description)\n        \n        # Find matching specialist\n        specialist = None\n        domain = None\n        \n        for cee_domain, spec in self.specialists.items():\n            if spec.handles_chapter(initial_chapter):\n                specialist = spec\n                domain = cee_domain\n                break\n        \n        if not specialist:\n            # Fallback to consumer goods for unknown chapters\n            specialist = self.specialists[CEEDomain.CONSUMER_MASS_MERCH]\n            domain = CEEDomain.CONSUMER_MASS_MERCH\n        \n        # Generate routing signals\n        routing_signals = self._extract_routing_signals(product_description, specialist)\n        \n        # Determine if escalation needed\n        escalation_required = self._needs_escalation(routing_signals, specialist)\n        qp_review_needed = self._needs_qp_review(initial_chapter, routing_signals)\n        \n        result = ChapterRoutingResult(\n            product_id=self._generate_product_id(product_description),\n            detected_chapter=initial_chapter,\n            cee_domain=domain,\n            specialist=specialist,\n            confidence=self._calculate_confidence(routing_signals, specialist),\n            routing_signals=routing_signals,\n            escalation_required=escalation_required,\n            qp_review_needed=qp_review_needed\n        )\n        \n        # Publish to pattern bus\n        if self.pattern_bus:\n            self.pattern_bus.publish(f\"chapter:route:{result.product_id}\", result)\n        \n        return result\n    \n    def get_chapter_signals(self, chapter: str) -> ChapterSignals:\n        \"\"\"Get classification signals to prioritize for given chapter\"\"\"\n        specialist = None\n        domain = None\n        \n        for cee_domain, spec in self.specialists.items():\n            if spec.handles_chapter(chapter):\n                specialist = spec\n                domain = cee_domain\n                break\n        \n        if not specialist:\n            specialist = self.specialists[CEEDomain.CONSUMER_MASS_MERCH]\n            domain = CEEDomain.CONSUMER_MASS_MERCH\n        \n        signals = ChapterSignals(\n            chapter=chapter,\n            domain=domain,\n            primary_signals=specialist.classification_signals[:2],  # Top 2 signals\n            secondary_signals=specialist.classification_signals[2:],  # Rest\n            pga_requirements=specialist.pga_agencies,\n            risk_screening=specialist.risk_flags\n        )\n        \n        # Publish to pattern bus\n        if self.pattern_bus:\n            self.pattern_bus.publish(f\"chapter:signals:{chapter}\", signals)\n        \n        return signals\n    \n    def process_jacks_intake(self, visual_data: Dict, coo_tags: List[str], \n                           price_points: Dict, upstream_materials: List[str]) -> ChapterRoutingResult:\n        \"\"\"Process Jack's 4 intake questions through CHINN routing\"\"\"\n        # Simulate product description from visual analysis\n        product_description = f\"Product type: {visual_data.get('product_type', 'unknown')}\"\n        \n        # Route through CHINN\n        routing_result = self.route_by_chapter(product_description)\n        \n        # Enhance with Jack's intake data\n        routing_result.routing_signals.update({\n            'visual_analysis': visual_data,\n            'country_tags': coo_tags,\n            'pricing': price_points,\n            'supply_chain': upstream_materials,\n            'jack_intake': True\n        })\n        \n        # Check if this falls under Jack's certification\n        jack_cert = self.qp_certifications.get(\"jack_chang\")\n        if jack_cert and jack_cert.is_certified_for_chapter(routing_result.detected_chapter):\n            routing_result.qp_review_needed = False  # Jack can handle\n            \n            # Publish QP pattern\n            if self.pattern_bus:\n                self.pattern_bus.publish(f\"chapter:qp:jack\", jack_cert)\n        \n        return routing_result\n    \n    def _detect_chapter(self, product_description: str) -> str:\n        \"\"\"Simple chapter detection based on keywords - can be enhanced with ML\"\"\"\n        keywords = product_description.lower()\n        \n        if any(word in keywords for word in ['shirt', 'dress', 'apparel', 'clothing', 'textile']):\n            return \"61\"\n        elif any(word in keywords for word in ['steel', 'iron', 'metal', 'aluminum']):\n            return \"72\"\n        elif any(word in keywords for word in ['computer', 'electronic', 'phone', 'device']):\n            return \"84\"\n        elif any(word in keywords for word in ['car', 'vehicle', 'automotive', 'engine']):\n            return \"87\"\n        elif any(word in keywords for word in ['chemical', 'pharmaceutical', 'drug']):\n            return \"29\"\n        elif any(word in keywords for word in ['food', 'meat', 'vegetable', 'fruit']):\n            return \"20\"\n        else:\n            return \"99\"  # Fallback\n    \n    def _extract_routing_signals(self, product_description: str, specialist: ChapterSpecialist) -> Dict[str, any]:\n        \"\"\"Extract routing signals based on specialist requirements\"\"\"\n        signals = {\n            'description': product_description,\n            'specialist_domain': specialist.domain.value,\n            'chapter_range': specialist.chapter_range,\n            'timestamp': datetime.now().isoformat()\n        }\n        \n        # Add domain-specific signal extraction\n        keywords = product_description.lower()\n        \n        for signal in specialist.classification_signals:\n            if signal in keywords or any(keyword in signal for keyword in keywords.split()):\n                signals[f'{signal}_detected'] = True\n        \n        return signals\n    \n    def _needs_escalation(self, signals: Dict[str, any], specialist: ChapterSpecialist) -> bool:\n        \"\"\"Determine if case needs escalation based on risk flags\"\"\"\n        for risk_flag in specialist.risk_flags:\n            if risk_flag.lower() in str(signals).lower():\n                return True\n        return False\n    \n    def _needs_qp_review(self, chapter: str, signals: Dict[str, any]) -> bool:\n        \"\"\"Determine if QP review is needed\"\"\"\n        # Always need QP review for complex cases or new chapters\n        complex_indicators = ['mixture', 'composite', 'multi-use', 'ambiguous']\n        description = str(signals.get('description', '')).lower()\n        \n        return any(indicator in description for indicator in complex_indicators)\n    \n    def _calculate_confidence(self, signals: Dict[str, any], specialist: ChapterSpecialist) -> float:\n        \"\"\"Calculate routing confidence score\"\"\"\n        base_confidence = 0.7\n        \n        # Boost confidence if specialist signals detected\n        detected_signals = sum(1 for key in signals.keys() if key.endswith('_detected'))\n        signal_boost = min(detected_signals * 0.1, 0.2)\n        \n        return min(base_confidence + signal_boost, 1.0)\n    \n    def _generate_product_id(self, description: str) -> str:\n        \"\"\"Generate unique product ID for tracking\"\"\"\n        import hashlib\n        return hashlib.md5(f\"{description}_{datetime.now()}\".encode()).hexdigest()[:12]\n\n# Global CHINN instance\nchinn = CHINNCore()\n\n# Convenience functions\ndef route_product(product_description: str, initial_chapter: Optional[str] = None) -> ChapterRoutingResult:\n    \"\"\"Route product to appropriate chapter specialist\"\"\"\n    return chinn.route_by_chapter(product_description, initial_chapter)\n\ndef get_signals_for_chapter(chapter: str) -> ChapterSignals:\n    \"\"\"Get classification signals for specific chapter\"\"\"\n    return chinn.get_chapter_signals(chapter)\n\ndef process_intake(visual_data: Dict, coo_tags: List[str], \n                  price_points: Dict, upstream_materials: List[str]) -> ChapterRoutingResult:\n    \"\"\"Process complete intake through CHINN\"\"\"\n    return chinn.process_jacks_intake(visual_data, coo_tags, price_points, upstream_materials)\n```\n\n## 2. Pattern Bus Integration\n\n```python\n# Add to oneworld_trade/core/pattern_bus.py - PatternType enum\nclass PatternType(Enum):\n    # ... existing patterns ...\n    CHAPTER = \"chapter\"  # Chapter routing decisions (from CHINN)\n```\n\n## 3. Unit Tests\n\n```python\n# tests/test_chinn.py\n\"\"\"\nUnit tests for CHINN - Chapter Hierarchy Intelligence Neural Network\n\"\"\"\n\nimport pytest\nfrom datetime import date\nfrom oneworld_trade.agents.chinn import (\n    CHINNCore, CEEDomain, ChapterSpecialist, QPCertification,\n    route_product, get_signals_for_chapter, process_intake\n)\n\nclass TestCHINNRouting:\n    \"\"\"Test CHINN routing functionality\"\"\"\n    \n    def setup_method(self):\n        self.chinn = CHINNCore()\n    \n    def test_ceramic_mug_routing(self):\n        \"\"\"Test ceramic mug routes to CEE008 Consumer Mass Merch\"\"\"\n        result = route_product(\"ceramic coffee mug with handle\")\n        \n        assert result.cee_domain == CEEDomain.CONSUMER_MASS_MERCH\n        assert result.detected_chapter in [\"69\", \"94\"]  # Ceramics or household items\n        assert result.confidence > 0.5\n    \n    def test_steel_beam_routing(self):\n        \"\"\"Test steel beam routes to CEE005 Base Metals\"\"\"\n        result = route_product(\"structural steel I-beam, grade A572\")\n        \n        assert result.cee_domain == CEEDomain.BASE_METALS\n        assert result.detected_chapter == \"72\"\n        assert \"AD_CVD\" in result.specialist.risk_flags\n        assert \"232_tariffs\" in result.specialist.risk_flags\n    \n    def test_silk_scarf_routing(self):\n        \"\"\"Test silk scarf routes to CEE004 Apparel/Textiles\"\"\"\n        result = route_product(\"100% silk scarf, woven, women's accessory\")\n        \n        assert result.cee_domain == CEEDomain.APPAREL_FOOTWEAR_TEXTILES\n        assert result.detected_chapter == \"61\"\n        assert \"fiber_content\" in result.specialist.classification_signals\n        assert \"UFLPA\" in result.specialist.risk_flags\n    \n    def test_smart_watch_disambiguation(self):\n        \"\"\"Test smart watch could route to electronics or consumer goods\"\"\"\n        result = route_product(\"smartwatch with fitness tracking and cellular connectivity\")\n        \n        # Should route to electronics due to \"electronic\" keyword\n        assert result.cee_domain in [CEEDomain.ELECTRONICS, CEEDomain.CONSUMER_MASS_MERCH]\n        assert result.detected_chapter == \"84\"  # Electronics\n        assert result.confidence > 0.6\n\nclass TestChapterSignals:\n    \"\"\"Test chapter-specific signal generation\"\"\"\n    \n    def test_textiles_signals(self):\n        \"\"\"Test textile chapter signals prioritize fiber content\"\"\"\n        signals = get_signals_for_chapter(\"61\")\n        \n        assert signals.domain == CEEDomain.APPAREL_FOOTWEAR_TEXTILES\n        assert \"fiber_content\" in signals.primary_signals\n        assert \"CPSC\" in signals.pga_requirements\n        assert \"UFLPA\" in signals.risk_screening\n    \n    def test_metals_signals(self):\n        \"\"\"Test metals chapter signals prioritize alloy grade\"\"\"\n        signals = get_signals_for_chapter(\"72\")\n        \n        assert signals.domain == CEEDomain.BASE_METALS\n        assert \"alloy_grade\" in signals.primary_signals\n        assert \"DOC\" in signals.pga_requirements\n        assert \"AD_CVD\" in signals.risk_screening\n\nclass TestQPCertification:\n    \"\"\"Test Qualified Person certification tracking\"\"\"\n    \n    def test_jack_certification(self):\n        \"\"\"Test Jack's QP certification for textiles\"\"\"\n        chinn = CHINNCore()\n        jack_cert = chinn.qp_certifications[\"jack_chang\"]\n        \n        assert jack_cert.qp_name == \"Jack Chang\"\n        assert jack_cert.is_certified_for_chapter(\"61\")\n        assert jack_cert.is_certified_for_chapter(\"62\") \n        assert not jack_cert.is_certified_for_chapter(\"84\")\n        assert jack_cert.cee_domain == CEEDomain.APPAREL_FOOTWEAR_TEXTILES\n    \n    def test_qp_routing_override(self):\n        \"\"\"Test QP routing overrides for certified chapters\"\"\"\n        result = route_product(\"cotton t-shirt, men's size large\")\n        \n        # Should detect Jack can handle this\n        assert result.detected_chapter == \"61\"\n        # QP review logic depends on complexity, not just chapter\n\nclass TestJackIntakeFlow:\n    \"\"\"Test Jack's 4-question intake workflow\"\"\"\n    \n    def test_complete_textile_intake(self):\n        \"\"\"Test complete textile intake through CHINN\"\"\"\n        visual_data = {\n            'product_type': 'shirt',\n            'construction': 'woven',\n            'gender': 'mens'\n        }\n        coo_tags = ['Made in Vietnam', 'Cotton from USA']\n        price_points = {'wholesale': 8.50, 'retail': 24.99}\n        upstream_materials = ['cotton yarn', 'polyester thread', 'plastic buttons']\n        \n        result = process_intake(visual_data, coo_tags, price_points, upstream_materials)\n        \n        assert result.cee_domain == CEEDomain.APPAREL_FOOTWEAR_TEXTILES\n        assert result.routing_signals['jack_intake'] == True\n        assert 'visual_analysis' in result.routing_signals\n        assert 'supply_chain' in result.routing_signals\n        \n        # Should route to Jack since he's certified for apparel\n        # (actual QP logic may vary based on complexity)\n\nclass TestSpecialistConfiguration:\n    \"\"\"Test specialist configuration and chapter mapping\"\"\"\n    \n    def test_cee_domain_coverage(self):\n        \"\"\"Test all CEE domains are configured\"\"\"\n        chinn = CHINNCore()\n        \n        assert len(chinn.specialists) == 10  # All 10 CEEs\n        \n        for domain in CEEDomain:\n            assert domain in chinn.specialists\n    \n    def test_chapter_range_no_gaps(self):\n        \"\"\"Test chapter ranges cover major HTS sections\"\"\"\n        chinn = CHINNCore()\n        \n        covered_chapters = set()\n        for specialist in chinn.specialists.values():\n            covered_chapters.update(specialist.chapter_range)\n        \n        # Should cover major chapters 1-99\n        assert \"01\" in covered_chapters or \"1\" in covered_chapters\n        assert \"61\" in covered_chapters  # Textiles\n        assert \"84\" in covered_chapters  # Electronics\n        assert \"87\" in covered_chapters  # Automotive\n    \n    def test_risk_flags_by_domain(self):\n        \"\"\"Test domain-specific risk flags are configured\"\"\"\n        chinn = CHINNCore()\n        \n        textiles = chinn.specialists[CEEDomain.APPAREL_FOOTWEAR_TEXTILES]\n        assert \"UFLPA\" in textiles.risk_flags\n        \n        metals = chinn.specialists[CEEDomain.BASE_METALS] \n        assert \"AD_CVD\" in metals.risk_flags\n        assert \"232_tariffs\" in metals.risk_flags\n        \n        electronics = chinn.specialists[CEEDomain.ELECTRONICS]\n        assert \"section_301\" in electronics.risk_flags\n        assert \"entity_list\" in electronics.risk_flags\n\nif __name__ == \"__main__\":\n    pytest.main([__file__])\n```\n\n## 4. Worked Example: Jack's Textile Intake\n\n```python\n# Example: Complete textile intake workflow\ndef demonstrate_jack_textile_intake():\n    \"\"\"\n    Worked example: Jack processes a textile import using CHINN\n    Product: Men's cotton dress shirts from Vietnam\n    \"\"\"\n    print(\"=== Jack's Textile Intake via CHINN ===\\n\")\n    \n    # Step 1: Visual/SKU Analysis\n    visual_data = {\n        'product_type': 'dress_shirt',\n        'construction': 'woven',\n        'gender': 'mens',\n        'collar_style': 'point_collar',\n        'sleeve_length': 'long_sleeve',\n        'closure': 'button_front'\n    }\n    print(f\"1. Visual Analysis: {visual_data}\")\n    \n    # Step 2: COO Tags\n    coo_tags = [\n        'Made in Vietnam',\n        'Cotton from USA', \n        'Buttons from China',\n        'Thread from Vietnam'\n    ]\n    print(f\"2. COO Tags: {coo_tags}\")\n    \n    # Step 3: Price Points\n    price_points = {\n        'fob_price': 12.50,\n        'wholesale_price': 18.75,\n        'retail_msrp': 49.99,\n        'currency': 'USD'\n    }\n    print(f\"3. Price Points: {price_points}\")\n    \n    # Step 4: Upstream Materials\n    upstream_materials = [\n        'cotton yarn (USA origin)',\n        'polyester thread (Vietnam)',\n        'plastic buttons (China)', \n        'interfacing (Vietnam)',\n        'labels (China)'\n    ]\n    print(f\"4. Upstream Materials: {upstream_materials}\")\n    \n    # Process through CHINN\n    print(f\"\\n=== CHINN Processing ===\")\n    \n    result = process_intake(visual_data, coo_tags, price_points, upstream_materials)\n    \n    print(f\"Detected Chapter: {result.detected_chapter}\")\n    print(f\"CEE Domain: {result.cee_domain}\")\n    print(f\"Confidence: {result.confidence:.2f}\")\n    print(f\"Escalation Required: {result.escalation_required}\")\n    print(f\"QP Review Needed: {result.qp_review_needed}\")\n    \n    # Show specialist signals\n    signals = get_signals_for_chapter(result.detected_chapter)\n    print(f\"\\nChapter {result.detected_chapter} Classification Signals:\")\n    print(f\"Primary: {signals.primary_signals}\")\n    print(f\"Secondary: {signals.secondary_signals}\")\n    print(f\"PGA Agencies: {signals.pga_requirements}\")\n    print(f\"Risk Screening: {signals.risk_screening}\")\n    \n    # Jack's QP Analysis\n    chinn = CHINNCore()\n    jack_cert = chinn.qp_certifications[\"jack_chang\"]\n    \n    print(f\"\\n=== Jack's QP Status ===\")\n    print(f\"Certified for Ch.{result.detected_chapter}: {jack_cert.is_certified_for_chapter(result.detected_chapter)}\")\n    print(f\"Jack's Domains: {jack_cert.certified_chapters}\")\n    print(f\"Certification Date: {jack_cert.certification_date}\")\n    \n    # Likely Classification Outcome\n    print(f\"\\n=== Likely Classification ===\")\n    print(f\"HTS: 6205.20.2020 (Men's cotton dress shirts, woven)\")\n    print(f\"Duty Rate: 19.7% (Column 1 General)\")\n    print(f\"Special Programs: Check Vietnam FTA eligibility\")\n    print(f\"Required Markings: Country of origin on shirt and packaging\")\n    print(f\"Fiber Disclosure: Required under FTC regulations\")\n    \n    # UFLPA Considerations\n    if \"UFLPA\" in result.specialist.risk_flags:\n        print(f\"\\n=== UFLPA Risk Assessment ===\")\n        print(f\"Cotton Origin: USA (LOW RISK)\")\n        print(f\"Manufacturing: Vietnam (LOW RISK)\")\n        print(f\"Recommendation: Maintain cotton origin documentation\")\n    \n    return result\n\nif __name__ == \"__main__\":\n    demonstrate_jack_textile_intake()\n```\n\nThis implementation provides:\n\n1. **Complete CHINN system** with all 10 CBP CEE domains mapped to HTS chapters\n2. **Chapter-specific routing** based on product analysis and domain expertise\n3. **QP certification tracking** starting with Jack's textile specialization\n4. **Pattern bus integration** for cross-INN communication\n5. **Jack's intake workflow** processing his 4 key questions through CHINN\n6. **Comprehensive tests** covering all major routing scenarios\n7. **Real-world example** showing complete textile import processing\n\nThe system routes classification requests to the appropriate domain expert, ensuring textiles get textile-specific analysis and metals get metals-specific analysis, just like CBP's actual CEE structure."
}