[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6F7KX52CDSZCNME4JTC6HZQ":3,"skill-same:01M6F7KX52CDSZCNME4JTC6HZQ":78},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":46,"linkOnly":47,"markdownContent":48,"markdownBody":49,"frontMatter":50,"refName":55,"commitSha":56,"skillDirectory":57,"forksCount":58,"repositoryHtmlUrl":59,"repositoryHomepageUrl":6,"ownerHtmlUrl":60,"ownerBlogUrl":61,"ownerDescription":62,"updatedAt":63,"install":64,"resources":77},"01M6F7KX52CDSZCNME4JTC6HZQ","langchain-data-handling",null,"Implement LangChain data privacy and handling best practices.\nUse when handling sensitive data, implementing PII protection,\nor ensuring data compliance in LLM applications.\nTrigger with phrases like \"langchain data privacy\", \"langchain PII\",\n\"langchain GDPR\", \"langchain data handling\", \"langchain compliance\".","Dicklesworthstone\u002Fpi_agent_rust","Dicklesworthstone","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F35050222?v=4","01M6CG26AR942CSVMN1M5E3DSE","Jeff Emanuel","01M6CG26EPP71PSHW1VDS71121",1992,"MIT","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Fplugins-community\u002Fplugins\u002Fsaas-packs\u002Flangchain-pack\u002Fskills\u002Flangchain-data-handling\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000033","数据分析师","PROFESSION","适用职业",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45],"langchain data handling","data","langchain","write","when","implementing","with","LLM Applications","trigger","io","protection","use","edit","implement","read","applications","pii","privacy","gdpr","data privacy","compliance","best practices","2026-11-10 10:12:30",false,"---\nname: langchain-data-handling\ndescription: |\n  Implement LangChain data privacy and handling best practices.\n  Use when handling sensitive data, implementing PII protection,\n  or ensuring data compliance in LLM applications.\n  Trigger with phrases like \"langchain data privacy\", \"langchain PII\",\n  \"langchain GDPR\", \"langchain data handling\", \"langchain compliance\".\nallowed-tools: Read, Write, Edit\nversion: 1.0.0\nlicense: MIT\nauthor: Jeremy Longshore \u003Cjeremy@intentsolutions.io>\n---\n\n# LangChain Data Handling\n\n## Overview\nBest practices for handling sensitive data, PII protection, and compliance in LangChain applications.\n\n## Prerequisites\n- Understanding of data privacy regulations (GDPR, CCPA)\n- LangChain application processing user data\n- Data classification framework\n\n## Instructions\n\n### Step 1: PII Detection and Masking\n```python\nimport re\nfrom typing import List, Tuple\nfrom dataclasses import dataclass\n\n@dataclass\nclass PIIPattern:\n    name: str\n    pattern: str\n    replacement: str\n\nPII_PATTERNS = [\n    PIIPattern(\"email\", r\"\\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\\.[A-Z|a-z]{2,}\\b\", \"[EMAIL]\"),\n    PIIPattern(\"phone\", r\"\\b\\d{3}[-.]?\\d{3}[-.]?\\d{4}\\b\", \"[PHONE]\"),\n    PIIPattern(\"ssn\", r\"\\b\\d{3}-\\d{2}-\\d{4}\\b\", \"[SSN]\"),\n    PIIPattern(\"credit_card\", r\"\\b\\d{4}[-\\s]?\\d{4}[-\\s]?\\d{4}[-\\s]?\\d{4}\\b\", \"[CREDIT_CARD]\"),\n    PIIPattern(\"ip_address\", r\"\\b\\d{1,3}\\.\\d{1,3}\\.\\d{1,3}\\.\\d{1,3}\\b\", \"[IP_ADDRESS]\"),\n    PIIPattern(\"date_of_birth\", r\"\\b\\d{1,2}\u002F\\d{1,2}\u002F\\d{2,4}\\b\", \"[DOB]\"),\n]\n\nclass PIIDetector:\n    \"\"\"Detect and mask PII in text.\"\"\"\n\n    def __init__(self, patterns: List[PIIPattern] = None):\n        self.patterns = patterns or PII_PATTERNS\n\n    def detect(self, text: str) -> List[Tuple[str, str, int, int]]:\n        \"\"\"Detect PII in text. Returns list of (type, value, start, end).\"\"\"\n        findings = []\n        for pattern in self.patterns:\n            for match in re.finditer(pattern.pattern, text, re.IGNORECASE):\n                findings.append((\n                    pattern.name,\n                    match.group(),\n                    match.start(),\n                    match.end()\n                ))\n        return findings\n\n    def mask(self, text: str) -> str:\n        \"\"\"Mask all PII in text.\"\"\"\n        masked = text\n        for pattern in self.patterns:\n            masked = re.sub(pattern.pattern, pattern.replacement, masked, flags=re.IGNORECASE)\n        return masked\n\n    def redact(self, text: str) -> Tuple[str, dict]:\n        \"\"\"Redact PII and return mapping for restoration.\"\"\"\n        redactions = {}\n        counter = {}\n\n        def replace(match, pattern_name, replacement):\n            count = counter.get(pattern_name, 0)\n            counter[pattern_name] = count + 1\n            key = f\"{replacement[1:-1]}_{count}\"\n            redactions[key] = match.group()\n            return f\"[{key}]\"\n\n        result = text\n        for pattern in self.patterns:\n            result = re.sub(\n                pattern.pattern,\n                lambda m, p=pattern: replace(m, p.name, p.replacement),\n                result,\n                flags=re.IGNORECASE\n            )\n\n        return result, redactions\n\n# Usage\ndetector = PIIDetector()\ntext = \"Contact john@example.com or call 555-123-4567\"\nmasked = detector.mask(text)\n# \"Contact [EMAIL] or call [PHONE]\"\n```\n\n### Step 2: Pre-processing Pipeline\n```python\nfrom langchain_core.runnables import RunnableLambda, RunnablePassthrough\n\ndef create_privacy_pipeline(chain):\n    \"\"\"Wrap chain with PII protection.\"\"\"\n    detector = PIIDetector()\n\n    def preprocess(input_data: dict) -> dict:\n        \"\"\"Mask PII before sending to LLM.\"\"\"\n        if \"input\" in input_data:\n            masked, redactions = detector.redact(input_data[\"input\"])\n            return {\n                **input_data,\n                \"input\": masked,\n                \"_redactions\": redactions\n            }\n        return input_data\n\n    def postprocess(output: str, redactions: dict = None) -> str:\n        \"\"\"Restore redacted values in output if needed.\"\"\"\n        # Note: Generally we DON'T restore PII in outputs\n        # This is just for cases where it's required\n        return output\n\n    privacy_chain = (\n        RunnableLambda(preprocess)\n        | chain\n    )\n\n    return privacy_chain\n\n# Usage\nfrom langchain_openai import ChatOpenAI\nfrom langchain_core.prompts import ChatPromptTemplate\n\nllm = ChatOpenAI(model=\"gpt-4o-mini\")\nprompt = ChatPromptTemplate.from_template(\"Summarize: {input}\")\nchain = prompt | llm\n\nsafe_chain = create_privacy_pipeline(chain)\nresult = safe_chain.invoke({\"input\": \"User john@example.com reported an issue\"})\n# LLM sees: \"User [EMAIL_0] reported an issue\"\n```\n\n### Step 3: Data Retention Policies\n```python\nfrom datetime import datetime, timedelta\nfrom typing import Optional\nimport hashlib\n\nclass DataRetentionManager:\n    \"\"\"Manage data retention for LLM interactions.\"\"\"\n\n    def __init__(self, retention_days: int = 30):\n        self.retention_days = retention_days\n        self.storage = {}  # Replace with actual storage\n\n    def store_interaction(\n        self,\n        user_id: str,\n        input_text: str,\n        output_text: str,\n        metadata: dict = None\n    ) -> str:\n        \"\"\"Store interaction with retention policy.\"\"\"\n        interaction_id = hashlib.sha256(\n            f\"{user_id}{datetime.now().isoformat()}{input_text}\".encode()\n        ).hexdigest()[:16]\n\n        # Mask PII before storage\n        detector = PIIDetector()\n\n        self.storage[interaction_id] = {\n            \"user_id_hash\": hashlib.sha256(user_id.encode()).hexdigest(),\n            \"input_masked\": detector.mask(input_text),\n            \"output_masked\": detector.mask(output_text),\n            \"created_at\": datetime.now().isoformat(),\n            \"expires_at\": (datetime.now() + timedelta(days=self.retention_days)).isoformat(),\n            \"metadata\": metadata or {}\n        }\n\n        return interaction_id\n\n    def cleanup_expired(self) -> int:\n        \"\"\"Remove expired interactions.\"\"\"\n        now = datetime.now()\n        expired = [\n            k for k, v in self.storage.items()\n            if datetime.fromisoformat(v[\"expires_at\"]) \u003C now\n        ]\n\n        for key in expired:\n            del self.storage[key]\n\n        return len(expired)\n\n    def delete_user_data(self, user_id: str) -> int:\n        \"\"\"GDPR right to erasure - delete all user data.\"\"\"\n        user_hash = hashlib.sha256(user_id.encode()).hexdigest()\n        to_delete = [\n            k for k, v in self.storage.items()\n            if v[\"user_id_hash\"] == user_hash\n        ]\n\n        for key in to_delete:\n            del self.storage[key]\n\n        return len(to_delete)\n```\n\n### Step 4: Consent Management\n```python\nfrom enum import Enum\nfrom pydantic import BaseModel\nfrom datetime import datetime\n\nclass ConsentType(str, Enum):\n    LLM_PROCESSING = \"llm_processing\"\n    DATA_RETENTION = \"data_retention\"\n    ANALYTICS = \"analytics\"\n    TRAINING = \"training\"  # For fine-tuning\n\nclass UserConsent(BaseModel):\n    user_id: str\n    consents: dict[ConsentType, bool]\n    updated_at: datetime\n    ip_address: str = None\n\nclass ConsentManager:\n    \"\"\"Manage user consent for data processing.\"\"\"\n\n    def __init__(self):\n        self.consents = {}\n\n    def set_consent(self, user_consent: UserConsent) -> None:\n        self.consents[user_consent.user_id] = user_consent\n\n    def check_consent(self, user_id: str, consent_type: ConsentType) -> bool:\n        \"\"\"Check if user has given consent.\"\"\"\n        if user_id not in self.consents:\n            return False\n        return self.consents[user_id].consents.get(consent_type, False)\n\n    def require_consent(self, consent_type: ConsentType):\n        \"\"\"Decorator to require consent before processing.\"\"\"\n        def decorator(func):\n            async def wrapper(user_id: str, *args, **kwargs):\n                if not self.check_consent(user_id, consent_type):\n                    raise PermissionError(\n                        f\"User {user_id} has not consented to {consent_type.value}\"\n                    )\n                return await func(user_id, *args, **kwargs)\n            return wrapper\n        return decorator\n\n# Usage\nconsent_manager = ConsentManager()\n\n@consent_manager.require_consent(ConsentType.LLM_PROCESSING)\nasync def process_with_llm(user_id: str, input_text: str):\n    return await chain.ainvoke({\"input\": input_text})\n```\n\n### Step 5: Audit Logging\n```python\nimport json\nfrom datetime import datetime\nfrom typing import Any\n\nclass AuditLogger:\n    \"\"\"Audit log for data access and processing.\"\"\"\n\n    def __init__(self, log_file: str = \"audit.jsonl\"):\n        self.log_file = log_file\n\n    def log(\n        self,\n        action: str,\n        user_id: str,\n        resource: str,\n        details: dict = None,\n        outcome: str = \"success\"\n    ) -> None:\n        \"\"\"Log an audit event.\"\"\"\n        event = {\n            \"timestamp\": datetime.now().isoformat(),\n            \"action\": action,\n            \"user_id_hash\": hashlib.sha256(user_id.encode()).hexdigest(),\n            \"resource\": resource,\n            \"outcome\": outcome,\n            \"details\": details or {}\n        }\n\n        with open(self.log_file, \"a\") as f:\n            f.write(json.dumps(event) + \"\\n\")\n\n    def log_llm_call(\n        self,\n        user_id: str,\n        model: str,\n        prompt_tokens: int,\n        has_pii: bool\n    ) -> None:\n        \"\"\"Log LLM API call.\"\"\"\n        self.log(\n            action=\"llm_call\",\n            user_id=user_id,\n            resource=f\"model\u002F{model}\",\n            details={\n                \"prompt_tokens\": prompt_tokens,\n                \"pii_detected\": has_pii\n            }\n        )\n\n# Callback for automatic audit logging\nclass AuditCallback(BaseCallbackHandler):\n    def __init__(self, audit_logger: AuditLogger, user_id: str):\n        self.audit_logger = audit_logger\n        self.user_id = user_id\n\n    def on_llm_end(self, response, **kwargs) -> None:\n        usage = response.llm_output.get(\"token_usage\", {}) if response.llm_output else {}\n        self.audit_logger.log_llm_call(\n            user_id=self.user_id,\n            model=response.llm_output.get(\"model_name\", \"unknown\") if response.llm_output else \"unknown\",\n            prompt_tokens=usage.get(\"prompt_tokens\", 0),\n            has_pii=False  # Set based on detection\n        )\n```\n\n## Data Handling Checklist\n- [ ] PII detection and masking implemented\n- [ ] Data retention policies defined\n- [ ] Consent management in place\n- [ ] Audit logging enabled\n- [ ] Right to erasure (GDPR) supported\n- [ ] Data minimization practiced\n- [ ] Encryption at rest and in transit\n\n## Resources\n- [GDPR Overview](https:\u002F\u002Fgdpr.eu\u002F)\n- [CCPA Compliance](https:\u002F\u002Foag.ca.gov\u002Fprivacy\u002Fccpa)\n- [OpenAI Data Usage Policy](https:\u002F\u002Fopenai.com\u002Fpolicies\u002Fapi-data-usage-policies)\n\n## Next Steps\nUse `langchain-security-basics` for additional security measures.\n","# LangChain Data Handling\n\n## Overview\nBest practices for handling sensitive data, PII protection, and compliance in LangChain applications.\n\n## Prerequisites\n- Understanding of data privacy regulations (GDPR, CCPA)\n- LangChain application processing user data\n- Data classification framework\n\n## Instructions\n\n### Step 1: PII Detection and Masking\n```python\nimport re\nfrom typing import List, Tuple\nfrom dataclasses import dataclass\n\n@dataclass\nclass PIIPattern:\n    name: str\n    pattern: str\n    replacement: str\n\nPII_PATTERNS = [\n    PIIPattern(\"email\", r\"\\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\\.[A-Z|a-z]{2,}\\b\", \"[EMAIL]\"),\n    PIIPattern(\"phone\", r\"\\b\\d{3}[-.]?\\d{3}[-.]?\\d{4}\\b\", \"[PHONE]\"),\n    PIIPattern(\"ssn\", r\"\\b\\d{3}-\\d{2}-\\d{4}\\b\", \"[SSN]\"),\n    PIIPattern(\"credit_card\", r\"\\b\\d{4}[-\\s]?\\d{4}[-\\s]?\\d{4}[-\\s]?\\d{4}\\b\", \"[CREDIT_CARD]\"),\n    PIIPattern(\"ip_address\", r\"\\b\\d{1,3}\\.\\d{1,3}\\.\\d{1,3}\\.\\d{1,3}\\b\", \"[IP_ADDRESS]\"),\n    PIIPattern(\"date_of_birth\", r\"\\b\\d{1,2}\u002F\\d{1,2}\u002F\\d{2,4}\\b\", \"[DOB]\"),\n]\n\nclass PIIDetector:\n    \"\"\"Detect and mask PII in text.\"\"\"\n\n    def __init__(self, patterns: List[PIIPattern] = None):\n        self.patterns = patterns or PII_PATTERNS\n\n    def detect(self, text: str) -> List[Tuple[str, str, int, int]]:\n        \"\"\"Detect PII in text. Returns list of (type, value, start, end).\"\"\"\n        findings = []\n        for pattern in self.patterns:\n            for match in re.finditer(pattern.pattern, text, re.IGNORECASE):\n                findings.append((\n                    pattern.name,\n                    match.group(),\n                    match.start(),\n                    match.end()\n                ))\n        return findings\n\n    def mask(self, text: str) -> str:\n        \"\"\"Mask all PII in text.\"\"\"\n        masked = text\n        for pattern in self.patterns:\n            masked = re.sub(pattern.pattern, pattern.replacement, masked, flags=re.IGNORECASE)\n        return masked\n\n    def redact(self, text: str) -> Tuple[str, dict]:\n        \"\"\"Redact PII and return mapping for restoration.\"\"\"\n        redactions = {}\n        counter = {}\n\n        def replace(match, pattern_name, replacement):\n            count = counter.get(pattern_name, 0)\n            counter[pattern_name] = count + 1\n            key = f\"{replacement[1:-1]}_{count}\"\n            redactions[key] = match.group()\n            return f\"[{key}]\"\n\n        result = text\n        for pattern in self.patterns:\n            result = re.sub(\n                pattern.pattern,\n                lambda m, p=pattern: replace(m, p.name, p.replacement),\n                result,\n                flags=re.IGNORECASE\n            )\n\n        return result, redactions\n\n# Usage\ndetector = PIIDetector()\ntext = \"Contact john@example.com or call 555-123-4567\"\nmasked = detector.mask(text)\n# \"Contact [EMAIL] or call [PHONE]\"\n```\n\n### Step 2: Pre-processing Pipeline\n```python\nfrom langchain_core.runnables import RunnableLambda, RunnablePassthrough\n\ndef create_privacy_pipeline(chain):\n    \"\"\"Wrap chain with PII protection.\"\"\"\n    detector = PIIDetector()\n\n    def preprocess(input_data: dict) -> dict:\n        \"\"\"Mask PII before sending to LLM.\"\"\"\n        if \"input\" in input_data:\n            masked, redactions = detector.redact(input_data[\"input\"])\n            return {\n                **input_data,\n                \"input\": masked,\n                \"_redactions\": redactions\n            }\n        return input_data\n\n    def postprocess(output: str, redactions: dict = None) -> str:\n        \"\"\"Restore redacted values in output if needed.\"\"\"\n        # Note: Generally we DON'T restore PII in outputs\n        # This is just for cases where it's required\n        return output\n\n    privacy_chain = (\n        RunnableLambda(preprocess)\n        | chain\n    )\n\n    return privacy_chain\n\n# Usage\nfrom langchain_openai import ChatOpenAI\nfrom langchain_core.prompts import ChatPromptTemplate\n\nllm = ChatOpenAI(model=\"gpt-4o-mini\")\nprompt = ChatPromptTemplate.from_template(\"Summarize: {input}\")\nchain = prompt | llm\n\nsafe_chain = create_privacy_pipeline(chain)\nresult = safe_chain.invoke({\"input\": \"User john@example.com reported an issue\"})\n# LLM sees: \"User [EMAIL_0] reported an issue\"\n```\n\n### Step 3: Data Retention Policies\n```python\nfrom datetime import datetime, timedelta\nfrom typing import Optional\nimport hashlib\n\nclass DataRetentionManager:\n    \"\"\"Manage data retention for LLM interactions.\"\"\"\n\n    def __init__(self, retention_days: int = 30):\n        self.retention_days = retention_days\n        self.storage = {}  # Replace with actual storage\n\n    def store_interaction(\n        self,\n        user_id: str,\n        input_text: str,\n        output_text: str,\n        metadata: dict = None\n    ) -> str:\n        \"\"\"Store interaction with retention policy.\"\"\"\n        interaction_id = hashlib.sha256(\n            f\"{user_id}{datetime.now().isoformat()}{input_text}\".encode()\n        ).hexdigest()[:16]\n\n        # Mask PII before storage\n        detector = PIIDetector()\n\n        self.storage[interaction_id] = {\n            \"user_id_hash\": hashlib.sha256(user_id.encode()).hexdigest(),\n            \"input_masked\": detector.mask(input_text),\n            \"output_masked\": detector.mask(output_text),\n            \"created_at\": datetime.now().isoformat(),\n            \"expires_at\": (datetime.now() + timedelta(days=self.retention_days)).isoformat(),\n            \"metadata\": metadata or {}\n        }\n\n        return interaction_id\n\n    def cleanup_expired(self) -> int:\n        \"\"\"Remove expired interactions.\"\"\"\n        now = datetime.now()\n        expired = [\n            k for k, v in self.storage.items()\n            if datetime.fromisoformat(v[\"expires_at\"]) \u003C now\n        ]\n\n        for key in expired:\n            del self.storage[key]\n\n        return len(expired)\n\n    def delete_user_data(self, user_id: str) -> int:\n        \"\"\"GDPR right to erasure - delete all user data.\"\"\"\n        user_hash = hashlib.sha256(user_id.encode()).hexdigest()\n        to_delete = [\n            k for k, v in self.storage.items()\n            if v[\"user_id_hash\"] == user_hash\n        ]\n\n        for key in to_delete:\n            del self.storage[key]\n\n        return len(to_delete)\n```\n\n### Step 4: Consent Management\n```python\nfrom enum import Enum\nfrom pydantic import BaseModel\nfrom datetime import datetime\n\nclass ConsentType(str, Enum):\n    LLM_PROCESSING = \"llm_processing\"\n    DATA_RETENTION = \"data_retention\"\n    ANALYTICS = \"analytics\"\n    TRAINING = \"training\"  # For fine-tuning\n\nclass UserConsent(BaseModel):\n    user_id: str\n    consents: dict[ConsentType, bool]\n    updated_at: datetime\n    ip_address: str = None\n\nclass ConsentManager:\n    \"\"\"Manage user consent for data processing.\"\"\"\n\n    def __init__(self):\n        self.consents = {}\n\n    def set_consent(self, user_consent: UserConsent) -> None:\n        self.consents[user_consent.user_id] = user_consent\n\n    def check_consent(self, user_id: str, consent_type: ConsentType) -> bool:\n        \"\"\"Check if user has given consent.\"\"\"\n        if user_id not in self.consents:\n            return False\n        return self.consents[user_id].consents.get(consent_type, False)\n\n    def require_consent(self, consent_type: ConsentType):\n        \"\"\"Decorator to require consent before processing.\"\"\"\n        def decorator(func):\n            async def wrapper(user_id: str, *args, **kwargs):\n                if not self.check_consent(user_id, consent_type):\n                    raise PermissionError(\n                        f\"User {user_id} has not consented to {consent_type.value}\"\n                    )\n                return await func(user_id, *args, **kwargs)\n            return wrapper\n        return decorator\n\n# Usage\nconsent_manager = ConsentManager()\n\n@consent_manager.require_consent(ConsentType.LLM_PROCESSING)\nasync def process_with_llm(user_id: str, input_text: str):\n    return await chain.ainvoke({\"input\": input_text})\n```\n\n### Step 5: Audit Logging\n```python\nimport json\nfrom datetime import datetime\nfrom typing import Any\n\nclass AuditLogger:\n    \"\"\"Audit log for data access and processing.\"\"\"\n\n    def __init__(self, log_file: str = \"audit.jsonl\"):\n        self.log_file = log_file\n\n    def log(\n        self,\n        action: str,\n        user_id: str,\n        resource: str,\n        details: dict = None,\n        outcome: str = \"success\"\n    ) -> None:\n        \"\"\"Log an audit event.\"\"\"\n        event = {\n            \"timestamp\": datetime.now().isoformat(),\n            \"action\": action,\n            \"user_id_hash\": hashlib.sha256(user_id.encode()).hexdigest(),\n            \"resource\": resource,\n            \"outcome\": outcome,\n            \"details\": details or {}\n        }\n\n        with open(self.log_file, \"a\") as f:\n            f.write(json.dumps(event) + \"\\n\")\n\n    def log_llm_call(\n        self,\n        user_id: str,\n        model: str,\n        prompt_tokens: int,\n        has_pii: bool\n    ) -> None:\n        \"\"\"Log LLM API call.\"\"\"\n        self.log(\n            action=\"llm_call\",\n            user_id=user_id,\n            resource=f\"model\u002F{model}\",\n            details={\n                \"prompt_tokens\": prompt_tokens,\n                \"pii_detected\": has_pii\n            }\n        )\n\n# Callback for automatic audit logging\nclass AuditCallback(BaseCallbackHandler):\n    def __init__(self, audit_logger: AuditLogger, user_id: str):\n        self.audit_logger = audit_logger\n        self.user_id = user_id\n\n    def on_llm_end(self, response, **kwargs) -> None:\n        usage = response.llm_output.get(\"token_usage\", {}) if response.llm_output else {}\n        self.audit_logger.log_llm_call(\n            user_id=self.user_id,\n            model=response.llm_output.get(\"model_name\", \"unknown\") if response.llm_output else \"unknown\",\n            prompt_tokens=usage.get(\"prompt_tokens\", 0),\n            has_pii=False  # Set based on detection\n        )\n```\n\n## Data Handling Checklist\n- [ ] PII detection and masking implemented\n- [ ] Data retention policies defined\n- [ ] Consent management in place\n- [ ] Audit logging enabled\n- [ ] Right to erasure (GDPR) supported\n- [ ] Data minimization practiced\n- [ ] Encryption at rest and in transit\n\n## Resources\n- [GDPR Overview](https:\u002F\u002Fgdpr.eu\u002F)\n- [CCPA Compliance](https:\u002F\u002Foag.ca.gov\u002Fprivacy\u002Fccpa)\n- [OpenAI Data Usage Policy](https:\u002F\u002Fopenai.com\u002Fpolicies\u002Fapi-data-usage-policies)\n\n## Next Steps\nUse `langchain-security-basics` for additional security measures.\n",{"name":5,"author":51,"license":15,"version":52,"description":53,"allowed-tools":54},"Jeremy Longshore \u003Cjeremy@intentsolutions.io>","1.0.0","Implement LangChain data privacy and handling best practices.\nUse when handling sensitive data, implementing PII protection,\nor ensuring data compliance in LLM applications.\nTrigger with phrases like \"langchain data privacy\", \"langchain PII\",\n\"langchain GDPR\", \"langchain data handling\", \"langchain compliance\".\n","Read, Write, Edit","main","90b50ed0778745a80fd5bfb13b1843d45b9e82f7","tests\u002Fext_conformance\u002Fartifacts\u002Fplugins-community\u002Fplugins\u002Fsaas-packs\u002Flangchain-pack\u002Fskills\u002Flangchain-data-handling",233,"https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust","https:\u002F\u002Fgithub.com\u002FDicklesworthstone","https:\u002F\u002Fwww.jeffreyemanuel.com\u002F","Building in NY","2026-11-15 01:54:15",{"source":8,"skillName":5,"commands":65,"prompt":75,"zipAvailable":76},[66,69,72],{"tool":67,"command":68},"npx","npx skills add Dicklesworthstone\u002Fpi_agent_rust --skill langchain-data-handling",{"tool":70,"command":71},"bunx","bunx skills add Dicklesworthstone\u002Fpi_agent_rust --skill langchain-data-handling",{"tool":73,"command":74},"pnpm","pnpm dlx skills add Dicklesworthstone\u002Fpi_agent_rust --skill langchain-data-handling","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Fplugins-community\u002Fplugins\u002Fsaas-packs\u002Flangchain-pack\u002Fskills\u002Flangchain-data-handling\u002FSKILL.md",true,[],[79,112,147,169,194,221],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":83,"licenseFileUrl":84,"sourceHtmlUrl":85,"categories":86,"keywords":92,"publishedAt":111},"01M6FCR7G5A005P364Z6ZWCJDP","theme-factory","Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors\u002Ffonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fcreative-design\u002Ftheme-factory\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fcreative-design\u002Ftheme-factory\u002FSKILL.md",[87],{"categoryNo":88,"categoryName":89,"taxonomyType":90,"taxonomyTypeDesc":91},"01JBSEEDCAT000000000000015","视觉设计","SCENARIO","使用场景",[93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,30,109,110],"theme factory","factory","theme","docs","toolkit","generate","artifacts","SET","fly","slides","landing","html","styling","artifact","pages","themes","you","fonts","2026-11-11 15:11:14",{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":83,"licenseFileUrl":116,"sourceHtmlUrl":117,"categories":118,"keywords":122,"publishedAt":46},"01M6FEGPPK0ZTKCCM0R3E3KS33","web-artifacts-builder","Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn\u002Fui). Use for complex artifacts requiring state management, routing, or shadcn\u002Fui components - not for simple single-file HTML\u002FJSX artifacts.","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Futilities\u002Fweb-artifacts-builder\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Futilities\u002Fweb-artifacts-builder\u002FSKILL.md",[119],{"categoryNo":120,"categoryName":121,"taxonomyType":90,"taxonomyTypeDesc":91},"01JBSEEDCAT000000000000009","前端开发",[123,124,99,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,35,104,140,141,142,143,144,145,146],"web artifacts builder","web","artifacts builder","builder","frontend","css","components","react","component","routing","state management","state","shadcn","file","ui","Tools","Tailwind","single-file","simple","ui components","multi","for","jsx","suite",{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":83,"licenseFileUrl":151,"sourceHtmlUrl":152,"categories":153,"keywords":158,"publishedAt":46},"01M6FEE9VTK1W4GY21AKREJXRG","webapp-testing","Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fsecurity\u002Fwebapp-testing\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fsecurity\u002Fwebapp-testing\u002FSKILL.md",[154,157],{"categoryNo":155,"categoryName":156,"taxonomyType":90,"taxonomyTypeDesc":91},"01JBSEEDCAT000000000000008","测试与质量",{"categoryNo":120,"categoryName":121,"taxonomyType":90,"taxonomyTypeDesc":91},[159,160,161,127,162,163,124,164,97,165,137,166,39,167,168,30,144],"testing","webapp testing","webapp","screenshots","playwright","debugging","browser","logs","behavior","local",{"skillNo":170,"name":171,"nameZh":6,"description":172,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":173,"sourceHtmlUrl":174,"categories":175,"keywords":176,"publishedAt":46},"01M6FE20F7ZTQBFWKF4MPJ2SBX","neuropixels-analysis","Neuropixels neural recording analysis. Load SpikeGLX\u002FOpenEphys data, preprocess, motion correction, Kilosort4 spike sorting, quality metrics, Allen\u002FIBL curation, AI-assisted visual analysis, for Neuropixels 1.0\u002F2.0 extracellular electrophysiology. Use when working with neural recordings, spike sorting, extracellular electrophysiology, or when the user mentions Neuropixels, SpikeGLX, Open Ephys, Kilosort, quality metrics, or unit curation.","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fscientific\u002Fneuropixels-analysis\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fscientific\u002Fneuropixels-analysis\u002FSKILL.md",[],[177,178,179,180,25,181,182,183,35,184,185,186,187,188,189,190,30,191,192,193,144,28],"analysis","neuropixels analysis","spike","metrics","open","quality","user","recordings","Neural","motion","visual","quality metrics","unit","sorting","curation","recording","load",{"skillNo":195,"name":196,"nameZh":6,"description":197,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":83,"licenseFileUrl":198,"sourceHtmlUrl":199,"categories":200,"keywords":204,"publishedAt":46},"01M6FDP0PDK6F1GGX395C7625Z","skill-creator","Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fproductivity\u002Fskill-creator\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fproductivity\u002Fskill-creator\u002FSKILL.md",[201],{"categoryNo":202,"categoryName":203,"taxonomyType":90,"taxonomyTypeDesc":91},"01JBSEEDCAT000000000000023","Skill与智能体开发",[205,206,207,208,209,210,211,212,213,214,215,216,217,218,30,144,219,28,220],"skill creator","creator","workflows","create","update","capabilities","knowledge","tool","integrations","specialized","guide","new skill","existing skill","users","Existing","new",{"skillNo":222,"name":223,"nameZh":6,"description":224,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":225,"sourceHtmlUrl":226,"categories":227,"keywords":229,"publishedAt":46},"01M6FDM8Y7VK7NYG9SM7QPA3PW","notebooklm","Use this skill to query your Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers from Gemini. Browser automation, library management, persistent auth. Drastically reduced hallucinations through document-only responses.","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fproductivity\u002Fnotebooklm\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FDicklesworthstone\u002Fpi_agent_rust\u002Fblob\u002Fmain\u002Ftests\u002Fext_conformance\u002Fartifacts\u002Ftemplates-davila7\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fproductivity\u002Fnotebooklm\u002FSKILL.md",[228],{"categoryNo":202,"categoryName":203,"taxonomyType":90,"taxonomyTypeDesc":91},[223,230,231,232,233,234,235,165,236,237,35,238,239,240,241,144,242],"google","browser automation","gemini","automation","auth","code","library","management","only","document","query","Citation","source"]