{"id":"ba406413-9d34-4f48-8fab-cdb431f918dc","entityType":"agent","slug":"clawhub-gechengling-insurance-agent-trainer","name":"Insurance Agent Intelligent Trainer","canonicalUrl":"https://www.xpersona.co/agent/clawhub-gechengling-insurance-agent-trainer","canonicalPath":"/agent/clawhub-gechengling-insurance-agent-trainer","generatedAt":"2026-10-10T03:02:14.696Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-09T23:27:51.973Z","emptyReason":null},"description":"AI-powered insurance agent training coach — auto-parses product docs, generates question banks, assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers predinig rate cut (3.0%) impact on sales scripts, new health insurance regulation, PIPL-compliant customer communication, and benchmark against AIA, Ping An, and Alibaba Cloud training systems. Keywords: 智能陪练, 代理人培训, 保险训练, 产品陪练, 话术训练, 角色扮演, 技能评估, AIA培训, 平安培训, 代理人展业, 保险销售, 客户面谈, 异议处理, 保险培训系统, 学习路径. Skill: Insurance Agent Intelligent Trainer Owner: gechengling Summary: AI-powered insurance agent training coach — auto-parses product docs, generates question banks, assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers predinig rate cut (3.0%) impact on sales scripts, new","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1.9K downloads reported by the source. Last updated 10/9/2026.","installCommand":"clawhub skill install s17ewqc4f2s6gpcbm88hy7fgvn85kg1g:insurance-agent-trainer","sourceUrl":"https://clawhub.ai/gechengling/insurance-agent-trainer","homepage":"https://clawhub.ai/gechengling/skills/insurance-agent-trainer","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/gechengling/insurance-agent-trainer","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/gechengling/skills/insurance-agent-trainer","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":66,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"AI-powered insurance agent training coach — auto-parses product docs, generates question banks, assesses agent skill levels (beginner/intermediate/advanced), sc"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-09T23:27:51.973Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T23:27:51.973Z","emptyReason":null},"stars":null,"forks":null,"downloads":1890,"packageName":null,"latestVersion":"5.2.5","tractionLabel":"1.9K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T23:27:51.972Z","emptyReason":null},"lastUpdatedAt":"2026-10-09T23:27:51.973Z","lastCrawledAt":"2026-10-09T23:27:51.972Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-10T23:27:51.972Z","lastVerifiedAt":null,"highlights":[{"version":"5.2.5","createdAt":"2026-09-24T05:17:12.726Z","changelog":"新增解析实操与评级示例、红线词扫雷与续期演练模式；环境/题库/模式表新增维度；修复未闭合代码块","fileCount":3,"zipByteSize":17609},{"version":"5.2.4","createdAt":"2026-08-29T14:05:56.627Z","changelog":"v5.2.4: Added second worked example in every module - critical illness product profile JSON, L3 expert trainee profile, a pension-planning question item, a health-insurance objection-cracking daily plan, and a role-play dialogue fragment. New table dimensions: weekly training hours per skill level, assessment focus per question category, recommended question count per difficulty, applicable scenario per training mode, evaluation method per metric, plus compliance red-line trigger rate. New question categories for pension planning and tax-advantaged insurance, new live-coaching training mode. Sales-environment regulatory table refreshed through 2026-08 (2.0% predetermined rate, marketing compliance, pension finance).","fileCount":3,"zipByteSize":14766},{"version":"5.2.3","createdAt":"2026-07-18T15:01:40.423Z","changelog":"v5.2.3: Fixed compliance metadata - removed under-scoped allowed-tools:[] and no-executable-code declarations. Updated capability declarations to reflect actual bundled scripts. Replaced security notice with honest capability notice. Refreshed regulatory news with July 2026 insurance reforms (分红险演示利率下调、适当性管理、65号文等).","fileCount":3,"zipByteSize":13010},{"version":"5.2.2","createdAt":"2026-06-28T13:14:18.763Z","changelog":"Insurance Agent Trainer v5.2.2 - Updated SKILL.md with the latest insurance regulatory news as of 2026-06-28, including key AI application guidance, regulatory work plans, and licensing changes. - Removed the file skill-card.md to streamline documentation. - No changes to capabilities or core architecture; informational updates only.","fileCount":10,"zipByteSize":41594},{"version":"5.2.1","createdAt":"2026-06-16T01:40:54.935Z","changelog":"- Security and compliance disclaimer section expanded and moved upward for greater visibility. - Internal system notices and methodology clarifications have been streamlined to reduce repetition. - Training and compliance warnings are now consolidated within the security notice, removing redundant statements elsewhere. - Obsolete file \"skill-card.md\" removed. No feature or content changes to core training logic or capabilities.","fileCount":10,"zipByteSize":41265},{"version":"5.2.0","createdAt":"2026-06-15T06:17:21.883Z","changelog":"**Insurance Agent Trainer 5.2.0 Changelog** - Added a new \"保险监管最新动态\" section with regulatory updates current to 2026-06-15, including new licensing regulations and recent enforcement trends. - Updated compliance and data security notices for clearer guidance. - Removed redundant file: _meta.json. - Incremented skill version to 5.2.0.","fileCount":10,"zipByteSize":41177},{"version":"5.1.3","createdAt":"2026-06-15T05:31:14.349Z","changelog":"Version 5.1.3 - Minor update to address review feedback and improve security declarations","fileCount":10,"zipByteSize":40873},{"version":"5.1.2","createdAt":"2026-06-04T15:18:23.107Z","changelog":"### Insurance Agent Trainer 5.1.2 Changelog - Clarified the status of the product document parser: now explicitly marked as a conceptual teaching demonstration, stating that no actual parsing or OCR is performed. - Updated explanatory notes to reinforce that workflow diagrams and parsing steps are illustrative only. - No functional changes or user-facing feature updates. All enhancements are for clearer compliance and expectation management.","fileCount":10,"zipByteSize":40838}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s17ewqc4f2s6gpcbm88hy7fgvn85kg1g:insurance-agent-trainer","setupComplexity":"low","setupSteps":["Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gechengling-insurance-agent-trainer/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gechengling-insurance-agent-trainer/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gechengling-insurance-agent-trainer/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-gechengling-insurance-agent-trainer/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-gechengling-insurance-agent-trainer/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-gechengling-insurance-agent-trainer/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-10T03:02:14.694Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gechengling-insurance-agent-trainer/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gechengling-insurance-agent-trainer/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gechengling-insurance-agent-trainer/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-gechengling-insurance-agent-trainer/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"high","updatedAt":"2026-10-09T23:27:51.973Z","emptyReason":null},"readme":"Skill: Insurance Agent Intelligent Trainer\n\nOwner: gechengling\n\nSummary: AI-powered insurance agent training coach — auto-parses product docs, generates question banks, assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers predinig rate cut (3.0%) impact on sales scripts, new health insurance regulation, PIPL-compliant customer communication, and benchmark against AIA, Ping An, and Alibaba Cloud training systems. Keywords: 智能陪练, 代理人培训, 保险训练, 产品陪练, 话术训练, 角色扮演, 技能评估, AIA培训, 平安培训, 代理人展业, 保险销售, 客户面谈, 异议处理, 保险培训系统, 学习路径.\n\nTags: CBIRC-compliance:1.0.0, L1-L3-competency:1.0.0, agent:5.2.4, agent-coaching:1.0.0, banking:5.0.0, china-insurance:1.0.0, dianjin:5.0.0, finance:5.0.0, insurance:5.2.4, insurance-agent:1.0.0, insurance-agent-trainer:5.2.5, insurance-sales:1.0.0, insurance-training:1.0.0, insurtech:1.0.0, interactive-learning:1.0.0, latest:5.2.5, life-insurance:1.0.0, objection-handling:1.0.0, personalized-training:1.0.0, product-knowledge:1.0.0, role-play:1.0.0, sales:5.2.4, sales-training:1.0.0, training:5.2.4, training-platform:1.0.0\n\nVersion history:\n\nv5.2.5 | 2026-09-24T05:17:12.726Z | user\n\n新增解析实操与评级示例、红线词扫雷与续期演练模式；环境/题库/模式表新增维度；修复未闭合代码块\n\nv5.2.4 | 2026-08-29T14:05:56.627Z | user\n\nv5.2.4: Added second worked example in every module - critical illness product profile JSON, L3 expert trainee profile, a pension-planning question item, a health-insurance objection-cracking daily plan, and a role-play dialogue fragment. New table dimensions: weekly training hours per skill level, assessment focus per question category, recommended question count per difficulty, applicable scenario per training mode, evaluation method per metric, plus compliance red-line trigger rate. New question categories for pension planning and tax-advantaged insurance, new live-coaching training mode. Sales-environment regulatory table refreshed through 2026-08 (2.0% predetermined rate, marketing compliance, pension finance).\n\nv5.2.3 | 2026-07-18T15:01:40.423Z | user\n\nv5.2.3: Fixed compliance metadata - removed under-scoped allowed-tools:[] and no-executable-code declarations. Updated capability declarations to reflect actual bundled scripts. Replaced security notice with honest capability notice. Refreshed regulatory news with July 2026 insurance reforms (分红险演示利率下调、适当性管理、65号文等).\n\nv5.2.2 | 2026-06-28T13:14:18.763Z | auto\n\nInsurance Agent Trainer v5.2.2\n\n- Updated SKILL.md with the latest insurance regulatory news as of 2026-06-28, including key AI application guidance, regulatory work plans, and licensing changes.\n- Removed the file skill-card.md to streamline documentation.\n- No changes to capabilities or core architecture; informational updates only.\n\nv5.2.1 | 2026-06-16T01:40:54.935Z | auto\n\n- Security and compliance disclaimer section expanded and moved upward for greater visibility.\n- Internal system notices and methodology clarifications have been streamlined to reduce repetition.\n- Training and compliance warnings are now consolidated within the security notice, removing redundant statements elsewhere.\n- Obsolete file \"skill-card.md\" removed. No feature or content changes to core training logic or capabilities.\n\nv5.2.0 | 2026-06-15T06:17:21.883Z | auto\n\n**Insurance Agent Trainer 5.2.0 Changelog**\n\n- Added a new \"保险监管最新动态\" section with regulatory updates current to 2026-06-15, including new licensing regulations and recent enforcement trends.\n- Updated compliance and data security notices for clearer guidance.\n- Removed redundant file: _meta.json.\n- Incremented skill version to 5.2.0.\n\nv5.1.3 | 2026-06-15T05:31:14.349Z | user\n\nVersion 5.1.3 - Minor update to address review feedback and improve security declarations\n\nv5.1.2 | 2026-06-04T15:18:23.107Z | auto\n\n### Insurance Agent Trainer 5.1.2 Changelog\n\n- Clarified the status of the product document parser: now explicitly marked as a conceptual teaching demonstration, stating that no actual parsing or OCR is performed.\n- Updated explanatory notes to reinforce that workflow diagrams and parsing steps are illustrative only.\n- No functional changes or user-facing feature updates. All enhancements are for clearer compliance and expectation management.\n\nv5.1.1 | 2026-06-04T15:09:53.751Z | auto\n\nInsurance Agent Intelligent Trainer v5.1.1\n\n- Strengthened security and legal disclaimers throughout documentation.\n- Clarified that all parsing, schedule, and profile functions are conceptual/educational frameworks, not executable or data-processing features.\n- Added explicit regulatory compliance reminders for sales scripts and objection handling.\n- Removed the sample skill-card.md file.\n\nv5.1.0 | 2026-06-03T14:34:45.276Z | auto\n\n**Skill insurance-agent-trainer v5.1.0 Changelog**\n\n- Updated version number to 5.1.0.\n- Added enhanced security and data protection disclaimers, especially clarifying no real OCR/product parsing and no data storage.\n- Revised trigger keyword rules: now requires explicit user confirmation before activation and does not auto-trigger on general training/coaching phrases.\n- Removed file: skill-card.md.\n- “allowed-tools” explicitly set as an empty list.\n\nv2.0.2 | 2026-06-02T15:20:23.869Z | auto\n\nInsurance Agent Trainer 2.0.2\n\n- Updated SKILL.md with enhanced data security disclaimers and user consent rules, clarifying educational scope and separating framework descriptions from real processing capabilities.\n- Improved trigger keyword logic: coaching functions now require explicit user confirmation before activation and will not trigger on generic training/coaching terms.\n- Added detailed warnings about personal information handling and compliance requirements for any real agent or client data.\n- Removed deprecated skill-card.md file for maintenance and clarity.\n\nv2.0.1 | 2026-06-02T14:37:42.221Z | auto\n\nInsurance Agent Intelligent Trainer 2.0.1\n\n- Clarified trigger activation rules for greater precision; general training/coaching keywords no longer trigger the skill.\n- Added a \"数据安全警告\" (Data Security Notice) section highlighting that the skill contains only framework descriptions, with no code or actual data processing.\n- Updated metadata fields, including the addition of `allowed-tools: []`.\n- Removed redundant file `skill-card.md` for better maintainability.\n- Documentation improvements throughout SKILL.md for consistency and compliance.\n\nv5.0.2 | 2026-06-01T22:35:13.865Z | user\n\nSecurity compliance: fixed garbled text, added capability declarations and advisory-only disclaimers; enriched content with detailed steps, rules, and report templates\n\nv5.0.1 | 2026-06-01T15:11:34.693Z | user\n\nSecurity compliance update: added capability declarations and advisory-only disclaimers to meet ClawHub security scan requirements\n\nv5.0.0 | 2026-05-31T02:11:50.096Z | user\n\n融合阿里点金（Dianjin）金融数字员工精髓，版本升级至5.0.0\n\nv4.0.1 | 2026-05-25T03:29:37.647Z | auto\n\n**Insurance Agent Intelligent Trainer v4.0.1 Changelog**\n\n- Added keyword slug and an extensive multilingual keywords list to the SKILL.md for improved discoverability and activation.\n- Updated regulatory insights: included 2026 insurance supervision changes such as new agent grading (four levels), product sales scope, mandatory risk assessment, and compliance requirements.\n- Minor edits and data refresh: clarified core capabilities, regulatory section updated for currency, and refined documentation structure.\n- Version info synchronized; removed outdated content.\n- No functional code or logic changes. Documentation-only update.\n\nv2.0.0 | 2026-05-11T06:09:08.987Z | auto\n\n**Version 2.0.0 — Major update with 2025-2026 regulatory and market changes**\n\n- Added coverage for 2025-2026 insurance sales environment, including 3.0% minimum rate, new health insurance regulations, agent exam updates, and PIPL-compliant customer communications.\n- Updated description and features to reflect response to recent market, product, and regulatory shifts.\n- Included a new section outlining key environmental changes and recommended sales script adjustments for agents.\n- Enhanced compliance and digital marketing coaching, especially for AI outbound and health insurance scenarios.\n- Maintained all core functionalities: document parsing, agent profiling, personalized training, and role-play drills.\n\nv1.0.0 | 2026-05-05T14:34:44.001Z | user\n\nInitial release: AI-powered insurance agent coaching system with (1) product document parser supporting PDF/Word/image OCR, (2) auto-generated question bank (116+ questions across 8 categories x 5 difficulty tiers), (3) 3-tier agent competency profiling (L1/L2/L3), (4) personalized daily training scheduler based on client visit schedules, (5) 5 interactive training modes (quick Q&A, role-play, case study, objection focus, assessment), (6) real-time progress tracking with radar chart. Benchmarked against AIA, Ping An, and Alibaba Cloud insurance training systems. References: question bank templates, agent profile framework, evaluation rubrics. Scripts: product parser, question generator, training scheduler.\n\nArchive index:\n\nArchive v5.2.5: 3 files, 17609 bytes\n\nFiles: skill-card.md (2444b), SKILL.md (38462b), _meta.json (142b)\n\nFile v5.2.5:SKILL.md\n\n---\nname: Insurance Agent Intelligent Trainer\ndescription: >\n AI-powered insurance agent training coach — auto-parses product docs, generates question banks,\n assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training\n based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers\n predinig rate cut (3.0%) impact on sales scripts, new health insurance regulation, PIPL-compliant\n customer communication, and benchmark against AIA, Ping An, and Alibaba Cloud training systems.\n Keywords: 智能陪练, 代理人培训, 保险训练, 产品陪练, 话术训练, 角色扮演, 技能评估, AIA培训, 平安培训, 代理人展业, 保险销售, 客户面谈, 异议处理, 保险培训系统, 学习路径.\nslug: insurance-agent-trainer\nversion: 5.2.5\nallowed-tools: []\ncapabilities:\n - educational-reference\n - advisory-only\n - requires-human-review\n - illustrative-code-samples\n---\n\n# Insurance Agent Intelligent Trainer / 保险代理人智能陪练系统\n\n> **⚠️ SECURITY NOTICE / 安全声明**\n> - **Type:** Educational reference / analytical framework ONLY\n> - **No runnable package is bundled:** the flow sketches, JSON samples and the short\n Python-style pseudocode below are illustrative reference material for you to adapt in\n your own environment; this skill ships no installer, service, parser or executable payload\n> - **No persistent storage, network calls, background execution, or credential collection**\n> - **All outputs are for reference only and require human review before real-world application**\n> - **This skill does NOT provide financial, legal, or insurance advice**\n> - **Users must exercise their own judgment and consult qualified professionals**\n>\n> **⚠️ 数据安全警告**\n> - 本技能仅提供保险代理人的培训辅导参考框架，**不内置可运行的解析器、题库引擎或存储服务**；\n 文中的流程示意、JSON 样例与伪代码均为教学参考，需由使用方在自己的系统中实现\n> - 所有文档解析、日程分析、画像评估的描述均为**教学参考框架**，**不包含实际的OCR或PDF解析引擎**\n> - 不会自动访问、存储或处理用户的任何培训数据或个人信息\n> - 培训计划和话术建议需结合用户实际业务场景调整，**不能替代专业培训师**\n> - **销售话术和异议处理仅为培训参考，实际使用须遵守《保险法》及相关监管规定，不得以AI输出替代合规审核**\n\n\n\n> **🔒 数据最小化前置声明（2026-09-24 新增，使用任何模块前先执行）**\n> 1. 只描述、不粘贴：客户信息以“45 岁私企老板、家庭年收入约 200 万”这类脱敏描述输入，禁止粘贴真实姓名、证件号、电话、住址、银行账号。\n> 2. 代理人档案同样脱敏：`agent_id` 用编号，姓名可用“张**”或化名。\n> 3. 训练记录默认只保留汇总指标（得分、弱项标签），不保留完整对话原文；确需保留须经本人同意并限定用途。\n> 4. 任何要写入文件或对外发送的训练报告、话术稿，先在对话中完整展示给用户预览，经明确确认后再落盘/发送。\n\n> **English:** AI-powered insurance agent coaching system — parses product documents, generates\n> personalized question banks, assesses agent competency levels, schedules daily training based on\n> client visits, and runs interactive role-play drills. Benchmarked against AIA, Ping An, and\n> Alibaba Cloud insurance training systems.\n>\n> **中文:** 保险代理人智能陪练系统——解析产品文档、自动生成问题库、评估代理人能力等级、\n> 结合当日客户拜访行程安排个性化训练、进行情景对练。对标友邦保险、平安保险、阿里云智能陪练水平。\n\n---\n\n## Trigger Keywords / 触发关键词\n\n**⚠️ 精确触发规则**：仅当用户明确提到保险代理人培训/陪练相关需求时激活。日常对话中提及\"培训\"、\"训练\"、\"coaching\"、\"agent training\"等通用词汇时**不会自动触发**。\n\n**用户确认规则**：当用户输入匹配以下关键词时，必须先确认用户意图：\n- \"您需要保险代理人陪练/培训服务吗？\"\n- 仅在用户明确确认后，才进入陪练模式\n\n激活关键词（需用户确认后生效）：\n\n- 保险陪练 / 产品陪练 / 智能陪练 / 代理人训练\n- 代理人培训 / 新人培训 / 保险话术训练\n- 产品演练 / 客户异议处理 / 保险销售训练\n- insurance agent training / insurance coaching / insurance product drill\n\n---\n\n## Core System Architecture / 核心系统架构\n\n### 0. 2025-2026 代理人销售环境最新变化（截至 2026-09-24）\n\n| 变化 | 内容 | 话术调整建议 | 对应培训模块 | 可信度标注 |\n|------|------|------------|\n| **预定利率降至3.0%** | 2024年9月后所有新产品执行 | 强调\"锁定3.0%长期确定收益\"，对比银行理财波动性 | 产品知识、促成话术 | 以官方最新发布为准 |\n| **分红险主导市场** | 分红险、万能险替代传统高利率产品 | 学会讲\"浮动收益+保底保障\"的双重价值 | 产品知识、异议处理 | 以官方最新发布为准 |\n| **健康险新规上线** | 2025年商业健康险管理办法修订 | 健康告知流程需更规范，禁止误导性说明 | 合规话术、情景对练 | 以官方最新发布为准 |\n| **代理人资格考试升级** | 2025年加入AI伦理、数字化服务模块 | 新人需补充数字化能力培训 | 综合考核、新人路径 | 以官方最新发布为准 |\n| **企微客户触达合规** | AI外呼需标注身份，营销需客户授权 | 培训合规营销话术，避免违规外呼 | 合规话术、直播带练 | 以官方最新发布为准 |\n| **预定利率进一步下调至2.0%** | 2026年监管引导普通型人身险预定利率上限降至2.0%，分红/万能演示利率同步压降 | 话术从\"锁定3.0%\"转为\"锁定2.0%长期确定+浮动分红对冲通胀\" | 产品知识、养老规划 | 以官方最新发布为准 |\n| **营销宣传合规强化** | 2026年整治\"炒停售\"\"夸大收益\"，自媒体/直播带货纳入监管 | 培训合规表达，禁用绝对化收益承诺与演示红线 | 合规话术、直播带练 | 以官方最新发布为准 |\n| **养老金融与税优扩容** | 个人养老金、商业养老金试点扩围，税优额度可期上调 | 强化养老规划与税优测算话术，绑定家庭现金流诊断 | 养老规划、税优保险 | 以官方最新发布为准 |\n| **报行合一与费用约束延续** | 渠道费用与手续费率约束延续，代理人收入结构从\"首年佣金\"向\"续期+服务\"倾斜 | 培训从\"单促成\"转向\"长期客户经营+续期服务\"话术，弱化一次性冲规模表述 | 促成话术、客户经营 | 以官方最新发布为准 |\n| **销售行为可回溯要求提高** | 双录、自媒体留痕、线上展业记录被更多用于检查 | 训练中同步演练\"可回溯动作\"：身份说明、条款重点提示、客户确认回执 | 合规话术、综合考核 | 以官方最新发布为准 |\n\n\n\n```\n┌─────────────────────────────────────────────────────────────────┐\n│ Insurance Agent Intelligent Trainer │\n├─────────────────────────────────────────────────────────────────┤\n│ ┌──────────────┐ ┌──────────────┐ ┌──────────────────────┐ │\n│ │ Product Doc │ │ Agent Profile│ │ Daily Schedule/Routes│ │\n│ │ Parser │ │ Engine │ │ Integration │ │\n│ │ (PDF/Word/ │ │ (Skill Level │ │ (Today's Visits & │ │\n│ │ Images) │ │ Assessment) │ │ Client Profiles) │ │\n│ └──────┬───────┘ └──────┬───────┘ └──────────┬───────────┘ │\n│ │ │ │ │\n│ ▼ ▼ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Question Bank Generation Engine │ │\n│ │ Product Knowledge │ Objection Handling │ Case Analysis │ │\n│ │ [5 difficulty tiers × 3 categories = 15 question types] │ │\n│ └──────────────────────────┬───────────────────────────────┘ │\n│ │ │\n│ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Personalized Training Scheduler │ │\n│ │ [Skill Level + Schedule + Product Priority = Daily Plan]│ │\n│ └──────────────────────────┬───────────────────────────────┘ │\n│ │ │\n│ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Interactive Training Engine │ │\n│ │ Role-play │ Real-time Feedback │ Progress Tracking │ │\n│ └──────────────────────────────────────────────────────────┘ │\n└─────────────────────────────────────────────────────────────────┘\n```\n\n---\n\n## Core Capabilities / 核心能力\n\n### 1. Product Document Parser / 产品文档解析引擎（教学演示）\n\n> **⚠️ 教学演示**：以下展示产品文档解析的**概念性教学方法论**，仅说明AI可如何辅助理解产品结构。**本技能不执行任何实际的PDF解析、OCR识别或文档提取操作。** 所有\"解析流程\"均为逻辑示意，实际应用需由具体的工程实现完成。\n\n**Supported formats (conceptual):** PDF, Word (.docx), scanned images (with OCR), plain text\n\n**Conceptual parsing pipeline (for reference):**\n\n```\nDocument Upload\n │\n ▼\n[Format Detection] → PDF / Word / Image / Text\n │\n ▼\n[Text Extraction] → Raw text content\n │\n ▼\n[Structure Analysis]\n ├─ Product name, type, target customers\n ├─ Coverage scope (death, medical, annuity, critical illness, etc.)\n ├─ Premium levels & payment periods\n ├─ Policy terms & exclusions\n ├─ Sales pitch key points\n ├─ Competitive advantages vs. similar products\n └─ Compliance notes & regulatory requirements\n │\n ▼\n[Structured Product Profile] → Ready for question generation\n```\n\n**Output: Structured Product Profile JSON**\n\n```json\n{\n \"product_name\": \"XX福享人生终身寿险(万能型)\",\n \"product_type\": \"whole-life insurance with universal account\",\n \"insurer\": \"国联人寿\",\n \"target_customers\": [\"30-50岁中高收入人群\", \"有财富传承需求\"],\n \"coverage\": {\n \"death_benefit\": \"100%-160%账户价值\",\n \"annuity_option\": \"60岁起可转换为年金\",\n \"waiver\": \"可选投保人保费豁免\"\n },\n \"premium\": {\n \"min_annual\": 12000,\n \"payment_periods\": [\"3年\", \"5年\", \"10年\", \"20年\"],\n \"min_coverage_years\": \"终身\"\n },\n \"key_selling_points\": [\n \"复利增值，万能账户历史结算利率4.5%-5.2%\",\n \"灵活追加，额外资金可随时进入万能账户\",\n \"身故保障与财富传承双重功能\"\n ],\n \"competitive_edges\": [\"结算利率优于同类竞品\", \"追加无上限\"],\n \"exclusions\": [\"投保人对被保险人的故意伤害\", \"2年内自杀(无民事行为能力人除外)\"],\n \"compliance_notes\": [\"需双录(录音录像)\", \"犹豫期15天\", \"等待期90天\"],\n \"difficulty_tags\": [\"新人友好\", \"需强化健康告知\", \"财务规划综合能力\"]\n}\n```\n\n**文档解析实操示例（2026-09-24 新增，示意）：**\n\n| 步骤 | 输入（示意） | 提取要点 | 易错点 |\n|------|-------------|---------|--------|\n| 1 识别产品边界 | 产品手册封面 + 条款首页 | 产品全称、条款编号、适用险种 | 把附加险当成主险提取 |\n| 2 抽取责任 | 保险责任章节 | 给付条件、给付比例、给付次数 | 漏掉\"同一事故仅给付一次\"限制 |\n| 3 抽取责任免除 | 责任免除章节 | 逐条列出，标注条款号 | 只摘抄标题不摘内容 |\n| 4 抽取费用与期间 | 费率表、条款附则 | 交费期间、等待期、犹豫期、宽限期 | 把等待期与犹豫期混淆 |\n| 5 形成卖点与风险 | 全篇 | 3 条卖点 + 2 条风险提示 | 卖点写成绝对化收益承诺 |\n\n**解析完的自检问题**：① 等待期、犹豫期、宽限期三个数字是否都取到了？② 责任免除是否逐条列出？\n③ 卖点里有没有\"保证\"\"必然\"\"最高\"这类红线词？\n\n**示例产品画像 2（重大疾病保险）：**\n```json\n{\n \"product_name\": \"XX康健终身重疾险(2026版)\",\n \"product_type\": \"critical illness insurance\",\n \"insurer\": \"国联人寿\",\n \"target_customers\": [\"28-50岁家庭经济支柱\", \"有重疾保障缺口人群\"],\n \"coverage\": {\n \"ci_types\": \"120种重疾+20种中症+40种轻症\",\n \"multiple_payout\": \"重疾1次+中症2次+轻症3次，累计最高260%保额\",\n \"death_benefit\": \"身故赔已交保费或现金价值较大者\"\n },\n \"premium\": {\n \"sample\": \"30岁男，50万保额，30年缴，年缴约 6800 元\",\n \"payment_periods\": [\"10年\",\"20年\",\"30年\"]\n },\n \"key_selling_points\": [\n \"重疾+中症+轻症三重递进保障\",\n \"轻中症豁免后续保费\",\n \"可附加恶性肿瘤二次赔付\"\n ],\n \"exclusions\": [\"投保前已患重疾\", \"遗传性疾病（条款约定）\", \"等待期内出险\"],\n \"compliance_notes\": [\"重疾定义以监管规范为准\", \"需明确告知等待期90-180天\", \"如实健康告知义务\"],\n \"difficulty_tags\": [\"健康告知敏感\", \"条款专业度高\", \"需结合医疗知识\"]\n}\n```\n\n---\n\n### 2. Agent Profile & Skill Assessment / 代理人画像与能力评估\n\n> **⚠️ 数据处理提醒**：以下代理人画像和日程数据为**演示示例**。实际使用时，用户应自行管理代理人数据的收集和存储，确保符合《个人信息保护法》及保险行业合规要求。请勿输入真实客户PII信息。\n\n**Three skill tiers:**\n\n| Tier | Level | Description | Training Focus | 建议训练时长/周 |\n|------|-------|-------------|----------------|----------------|\n| 🌱 **L1 - 入门级** | Beginner | < 1 year experience, struggles with product details and objection handling | Foundation: product knowledge, basic sales scripts, simple objection responses | 5-8 小时（晨会快练+情景对练） |\n| ⚡ **L2 - 进阶级** | Intermediate | 1-3 years, solid product knowledge but inconsistent closing rate | Application: complex scenarios, multi-product combination, competitive replacement, high-net-worth clients | 3-5 小时（聚焦弱项情景对练） |\n| 🎯 **L3 - 专家级** | Advanced | 3+ years, high performance, needs strategy for complex cases | Mastery: enterprise/group clients, tax planning, estate planning, competitive stealing, mentoring skills | 2-3 小时（策略复盘+带教新人） |\n\n**Profile structure:**\n\n```json\n{\n \"agent_id\": \"AG20240001\",\n \"name\": \"张明\",\n \"level\": \"L2\",\n \"level_label\": \"进阶级\",\n \"tenure_years\": 2.5,\n \"certifications\": [\"保险代理人资格证\", \"健康险销售资质\"],\n \"performance\": {\n \"monthly_premium_target\": 50000,\n \"monthly_premium_actual\": 42000,\n \"closing_rate\": 0.32,\n \"avg_policy_size\": 18500,\n \"new_customer_rate\": 0.45\n },\n \"product_mastery\": {\n \"term_life\": 0.85,\n \"whole_life\": 0.72,\n \"critical_illness\": 0.58,\n \"medical_insurance\": 0.80,\n \"annuity\": 0.45,\n \"investment_linked\": 0.38\n },\n \"weak_points\": [\n \"健康险异议处理不够熟练\",\n \"不了解高端客户的税务筹划需求\",\n \"组合产品销售话术单一\"\n ],\n \"strong_points\": [\n \"老客户维护能力强\",\n \"缘故市场开拓优秀\"\n ],\n \"daily_schedule\": [\n {\"time\": \"09:00-10:00\", \"activity\": \"晨会\", \"location\": \"营业部\"},\n {\"time\": \"10:30-12:00\", \"activity\": \"拜访客户A（国企中层，有养老需求）\", \"location\": \"客户公司\"},\n {\"time\": \"14:00-15:30\", \"activity\": \"拜访客户B（私企业主，健康险需求）\", \"location\": \"客户公司\"},\n {\"time\": \"16:00-17:30\", \"activity\": \"缘故客户C（教育金规划）\", \"location\": \"咖啡厅\"}\n ]\n}\n```\n\n**能力等级评定示例（2026-09-24 新增）：**\n\n| 代理人 | 工龄 | 产品掌握度均值 | 成交率 | 弱项标签 | 评定等级 | 依据 |\n|--------|------|---------------|--------|---------|---------|------|\n| 示例A | 8 个月 | 0.62 | 0.18 | 条款细节、健康告知 | L1 入门级 | 工龄<1 年且产品掌握度均值<0.70 |\n| 示例B | 2.5 年 | 0.70 | 0.32 | 组合销售、年金 | L2 进阶级 | 产品知识扎实但成交率波动大 |\n| 示例C | 6 年 | 0.90 | 0.48 | 复杂传承架构 | L3 专家级 | 高成交率 + 高净值客户经营能力 |\n\n**评级使用注意**：等级只用于安排训练强度，不用于人事考核；同一代理人不同产品线可落在不同等级，\n建议按产品线条分别记录，避免用一个总分掩盖结构性短板。\n\n**示例画像 2（L3 专家级）：**\n```json\n{\n \"agent_id\": \"AG20230088\",\n \"name\": \"李华\",\n \"level\": \"L3\",\n \"level_label\": \"专家级\",\n \"tenure_years\": 6,\n \"certifications\": [\"保险代理人资格证\", \"CFP国际金融理财师\", \"私人银行家\"],\n \"performance\": {\n \"monthly_premium_target\": 200000,\n \"monthly_premium_actual\": 235000,\n \"closing_rate\": 0.48,\n \"avg_policy_size\": 86000,\n \"new_customer_rate\": 0.62\n },\n \"product_mastery\": {\n \"term_life\": 0.95, \"whole_life\": 0.92, \"critical_illness\": 0.90,\n \"medical_insurance\": 0.93, \"annuity\": 0.88, \"investment_linked\": 0.82\n },\n \"weak_points\": [\"家族信托等复杂传承架构经验不足\", \"跨境税务筹划需外部专家协同\"],\n \"strong_points\": [\"高净值客户经营\", \"企业团险开拓\", \"复杂方案设计\"],\n \"coaching_focus\": [\"传承架构进阶\", \"监管合规红线强化\", \"带教新人方法论\"]\n}\n```\n\n---\n\n### 3. Question Bank Generation / 问题库自动生成（教学模板）\n\n> **⚠️ 教学演示**：以下问题库和话术为**培训场景的教学参考模板**，展示如何结构化设计代理人训练内容。所有涉及销售话术、竞品对比、异议处理的内容均为**培训素材**，实际销售行为须遵循《保险法》及相关监管规定，并经持牌保险专业人士审核后方可执行。\n\n**Generated from product profile + agent level + training objectives**\n\n#### Question Types (15 categories across 3 dimensions)\n\n**By Category:**\n\n| Category | Description | Example | 考核重点 | 常见失分点 |\n|----------|-------------|---------|---------|\n| **产品知识** | Product features, terms, coverage | \"XX福的等待期是多久？\" | 条款准确性、关键利益点无误 | 把等待期答成犹豫期；漏讲除外责任 |\n| **客户画像** | Target customer identification | \"什么样的客户适合购买这款产品？\" | 需求诊断与匹配逻辑 |\n| **异议处理** | Objection handling scripts | \"客户说'我已经有社保了，不需要商业保险'，如何回应？\" | 共情+数据化反驳能力 | 直接否定客户、用恐吓式话术 |\n| **案例分析** | Real case discussion | \"40岁国企中层，年薪50万，如何用这款产品做养老规划？\" | 方案完整性与定制化 |\n| **合规话术** | Compliance-approved scripts | \"如何向客户解释犹豫期和退保损失？\" | 红线词零触发 | 使用\"保本\"\"稳赚\"\"最高收益\"等表述 |\n| **竞品对比** | vs. competitors | \"相比平安福，这款产品的核心优势是什么？\" | 客观不贬损竞品 |\n| **促成话术** | Closing techniques | \"客户表现出购买意向，如何自然促成？\" | 时机把握自然度 |\n| **交叉销售** | Multi-product combination | \"如何将主险与医疗险组合销售？\" | 保障缺口覆盖度 |\n| **养老规划** | 养老现金流与替代率测算 | \"客户55岁期望退休月领8000，如何测算缺口？\" | 测算逻辑与工具使用 |\n| **税优保险** | 个人养老金/税优健康险政策应用 | \"年缴1.2万养老金，节税多少？\" | 政策准确、不夸大节税 |\n| **客户经营** | 续期服务与转介绍 | \"成交一年后的客户，如何自然开启加保话题？\" | 服务触点设计、不打扰式回访 |\n| **数字化展业** | 企微/自媒体合规触达 | \"在朋友圈讲产品，哪些表述不能出现？\" | 平台规则与监管红线双重合规 |\n\n**By Difficulty (5 tiers):**\n\n| Level | Target Audience | Question Complexity | 建议题量/次 |\n|--------|----------------|---------------------|------------|\n| ⭐ 基础 | L1新人 | 单一产品，单一问题，直接答案 | 10-15 题 |\n| ⭐⭐ 入门 | L1-L2 | 单一产品，1-2个知识点，需要解释 | 15-20 题 |\n| ⭐⭐⭐ 进阶 | L2 | 单一产品，3-5个知识点，需组合分析 | 20-30 题 |\n| ⭐⭐⭐⭐ 高阶 | L2-L3 | 多产品组合，竞争替换，高净值客户 | 25-35 题 |\n| ⭐⭐⭐⭐⭐ 专家 | L3 | 综合方案，税务筹划，财富传承 | 30-40 题 |\n\n#### Question Bank Generation Prompt:\n\n```\nBased on the product profile provided, generate a question bank with:\n\n1. For each difficulty tier (基础/入门/进阶/高阶/专家):\n - 5 multiple choice questions (产品知识)\n - 3 case analysis questions\n - 3 objection handling scenarios\n - 2 competitive comparison questions\n - 1 closing technique exercise\n\n2. Total: 65+ questions per product\n\n3. For each question, provide:\n - Question text\n - Difficulty level (1-5)\n - Category (产品知识/异议处理/案例分析/竞品对比/促成话术)\n - Ideal answer / model response\n - Evaluation criteria (excellent/good/needs-improvement)\n - Coaching tips for the trainer\n```\n\n**示例生成题目（养老规划类别 / ⭐⭐⭐ 进阶）：**\n- **题目**：客户 55 岁，当前社保养老金预计月领 3500 元，期望退休后月生活支出 8000 元，如何测算商业养老金缺口？\n- **参考答案**：缺口 = (8000 - 3500) × 12 × 退休年限（按 25 年计）≈ 135 万；结合预期投资收益率反推年缴/趸交金额，并叠加通胀与医疗支出弹性。\n- **评分**：优秀（准确测算+工具使用）/ 良好（逻辑正确但忽略通胀）/ 待改进（未考虑长寿风险）。\n- **教练提示**：引导代理人用\"替代率\"概念切入，避免直接推销产品。\n\n---\n\n### 4. Personalized Training Scheduler / 个性化训练调度引擎（方法论演示）\n\n> **⚠️ 教学演示**：以下调度算法、代理人画像及日程数据均为**教学方法论的概念性展示**。**本技能不实际采集、存储或处理任何代理人或客户数据**。所有姓名、日程、业绩数据均为虚构示例，仅用于说明逻辑框架。\n\n**Input factors:**\n\n```\nAgent Profile (Level + Weak Points)\n +\nToday's Client Schedule (Who → What need → What product)\n +\nProduct Priority Matrix\n =\nPersonalized Daily Training Plan\n```\n\n**Scheduling Algorithm:**\n\n```python\ndef generate_daily_training_plan(agent_profile, daily_schedule, products):\n \"\"\"\n Generate personalized training plan for the day.\n \"\"\"\n # Step 1: Identify today's client visit products\n today_products = extract_products_from_schedule(daily_schedule)\n \n # Step 2: Get agent's weakness areas for these products\n weakness_map = get_weakness_for_products(\n agent_profile.weak_points, \n today_products\n )\n \n # Step 3: Calculate training time available\n available_minutes = calculate_available_training_time(daily_schedule)\n \n # Step 4: Prioritize by impact × weakness × product value\n training_queue = prioritize_training(\n weakness_map,\n today_products,\n agent_profile.level,\n time_constraint=available_minutes\n )\n \n # Step 5: Generate session plan\n sessions = split_into_sessions(training_queue, available_minutes)\n \n return {\n \"date\": today,\n \"agent\": agent_profile.name,\n \"total_minutes\": available_minutes,\n \"sessions\": sessions,\n \"focus_products\": today_products,\n \"key_objectives\": get_key_objectives(training_queue)\n }\n```\n\n**Example Daily Training Plan:**\n\n```json\n{\n \"date\": \"2026-05-05\",\n \"agent\": \"张明\",\n \"level\": \"L2\",\n \"total_minutes\": 90,\n \"sessions\": [\n {\n \"time\": \"08:00-08:20\",\n \"duration\": 20,\n \"type\": \"晨间快练\",\n \"mode\": \"快问快答\",\n \"focus\": \"年金险产品知识（高频问题5题）\",\n \"product\": \"福享人生终身寿险\",\n \"objective\": \"巩固年金转换权的计算逻辑\"\n },\n {\n \"time\": \"12:30-13:00\",\n \"duration\": 30,\n \"type\": \"午间强化\",\n \"mode\": \"情景对练\",\n \"focus\": \"健康险异议处理\",\n \"scenario\": \"客户：\"我有社保，不需要商业医疗险\"\",\n \"product\": \"康健医疗保险\",\n \"level\": \"⭐⭐⭐ 进阶\",\n \"coaching_tips\": \"引导客户认识到社保报销比例上限，用自费药比例对比引发需求\"\n },\n {\n \"time\": \"17:30-18:30\",\n \"duration\": 40,\n \"type\": \"晚间复盘\",\n \"mode\": \"案例分析 + 角色扮演\",\n \"focus\": \"私企业主综合保障方案\",\n \"scenario\": \"45岁私企老板，年收入200万，已有多份保单，如何做加保方案？\",\n \"products\": [\"终身寿险+万能账户\", \"高端医疗\", \"企业财产险\"],\n \"level\": \"⭐⭐⭐⭐ 高阶\",\n \"model_response_guide\": \"从家庭资产与企业资产隔离角度切入，引出终身寿险的债务隔离和传承功能\"\n }\n ],\n \"key_metrics_to_track\": [\n \"异议处理响应时间（目标<30秒）\",\n \"产品知识点正确率（目标>85%）\",\n \"方案组合完整性（3单以上产品覆盖）\"\n ]\n}\n```\n\n**示例计划 2（健康险异议攻坚日）：**\n```json\n{\n \"date\": \"2026-08-12\",\n \"agent\": \"李华\",\n \"level\": \"L3\",\n \"total_minutes\": 60,\n \"sessions\": [\n {\n \"time\": \"13:00-13:20\",\n \"duration\": 20,\n \"type\": \"午间强化\",\n \"mode\": \"异议攻关\",\n \"focus\": \"健康险'已有社保'高频异议\",\n \"scenario\": \"客户：'我有医保，重疾险没必要'\",\n \"level\": \"⭐⭐⭐ 进阶\",\n \"coaching_tips\": \"用'医保目录外用药+收入补偿'双轴拆解，量化缺口而非否定客户\"\n },\n {\n \"time\": \"18:00-18:40\",\n \"duration\": 40,\n \"type\": \"晚间复盘\",\n \"mode\": \"案例研讨 + 角色扮演\",\n \"focus\": \"高净值客户重疾+医疗+寿险组合\",\n \"scenario\": \"50岁企业主，家庭年收入300万，已有多张保单如何查漏补缺？\",\n \"level\": \"⭐⭐⭐⭐ 高阶\",\n \"model_response_guide\": \"从企业资产与家庭资产隔离、重疾收入补偿、医疗高端资源三维度切入\"\n }\n ],\n \"key_metrics_to_track\": [\n \"异议处理响应时间（目标<25秒）\",\n \"方案组合维度（目标≥4个）\",\n \"合规红线触发（目标0次）\"\n ]\n}\n```\n\n---\n\n### 5. Interactive Training Session / 智能陪练对话引擎\n\n**Session modes:**\n\n| Mode | Description | Duration | Best For | 适用场景 | 教练干预时机 |\n|------|-------------|----------|----------|---------|\n| **快问快答** | Rapid-fire Q&A | 5-10 min | Pre-meeting warmup | 晨会热身、拜访前激活产品知识 | 连续 2 题答错即给提示 |\n| **情景对练** | Role-play (client vs. agent) | 15-30 min | Skill practice | 健康险/养老险高频异议实战 | 出现红线词立即打断纠正 |\n| **案例研讨** | Real case analysis | 20-40 min | Advanced agents | 高净值综合保障方案设计 | 方案缺口遗漏 ≥2 项时给结构化提示 |\n| **异议攻关** | Objection busting focus | 10-15 min | Weak point training | 单一弱项（如\"已有社保\"）专项突破 |\n| **综合考核** | Full simulation exam | 30-60 min | Level assessment | 晋升/季度能力认证 |\n| **直播带练** | 模拟自媒体/直播讲保险 | 15-25 min | 数字化展业 | 合规表达与镜头前讲产品 | 出现绝对化表述当场叫停并重录 |\n| **续期服务演练** | 成交后回访与加保触达 | 10-20 min | 客户经营 | 续期提醒、理赔陪伴、转介绍开场 | 出现\"推销感\"过强时提示改为服务切入 |\n| **红线词扫雷** | 集中训练合规表达 | 8-12 min | 全员必修 | 演示利益、收益承诺、停售炒作 | 命中即判不合格并给正确说法 |\n\n**Real-time coaching during training:**\n\n```\nAgent Response\n │\n ▼\n[Natural Language Understanding] → Extract key claims, tone, strategy\n │\n ▼\n[Evaluation Engine]\n ├─ Product knowledge accuracy ✓/✗\n ├─ Objection handling effectiveness (1-5)\n ├─ Compliance adherence ✓/✗\n ├─ Closing attempt timing (good/early/late/missing)\n ├─ Client empathy signals ✓/✗\n └─ Product combination logic ✓/✗\n │\n ▼\n[Real-time Coaching Feedback]\n ├─ Immediate tip (if struggling): \"💡 提示：可以先问客户目前的保障缺口...\"\n ├─ Completion praise (if excellent): \"🌟 完美！您已经很好地识别了客户需求\"\n └─ Post-question summary: \"本轮得分 85/100。建议加强：竞品对比环节\"\n```\n\n**Training session flow:**\n\n```\n1. 导入 (5%) → 介绍训练目标和产品背景\n2. 暖场 (10%) → 快问快答热身，激活产品知识\n3. 主体 (60%) → 情景对练：客户角色扮演 + 实时点评\n4. 复盘 (20%) → AI给出详细反馈：优点/不足/改进建议\n5. 行动 (5%) → 下次拜访的具体行动计划\n```\n\n**示例对练对话片段（健康险异议攻关）：**\n```\nAI(客户): \"我单位福利好，重疾险真没必要买。\"\nAgent: \"您说的对，单位福利是重要保障。不过重疾理赔是'确诊即付'的一笔钱——您想过没有，万一需要长期康复，单位会不会照发全额工资？\"\nAI(客户): \"那倒不会，病假工资大概只发底薪……\"\nAgent: \"这就是缺口。重疾险补的正是'收入中断+自费药'这两块。我们按您月支出算一下具体差额？\"\n[实时点评] ✅ 共情到位；✅ 用'收入补偿'替代'恐吓式'话术；⚠️ 下一步应主动给出测算而非直接推产品。\n```\n\n---\n\n### 6. Effect Assessment & Progress Tracking / 效果评估与进度追踪\n\n**Metrics tracked per session:**\n\n| Metric | Definition | Target | 评估方式 | 提升抓手 |\n|--------|------------|--------|---------|\n| **产品知识得分** | 知识点正确率 | L1: ≥70%, L2: ≥80%, L3: ≥90% | 自动判分题库 + 人工抽检 | 错题回到条款原文定位，不靠背话术 |\n| **异议处理时效** | 从异议提出到满意回答的时间 | < 30秒 | 对话时间戳测算 | 建\"异议—回应\"对照卡，形成肌肉记忆 |\n| **促成成功率** | 能否自然引入促成信号 | ≥ 1次有效尝试 | 教练引擎标记 + 主管复核 |\n| **话术合规率** | 合规敏感词使用正确性 | 100% | 合规词库实时监测 |\n| **方案完整性** | 保障覆盖广度 | ≥ 3个维度 | 结构化评分表 |\n| **合规红线触发率** | 触碰禁语/误导表述次数 | 0 次 | 实时拦截 + 事后复盘 | 红线词清单前置到每次对练开场 |\n\n**Progress report structure:**\n\n```markdown\n## 📊 代理人张明 训练报告 - 2026-05-05\n\n### 综合得分: ⭐⭐⭐⭐ (78/100)\n\n| 维度 | 本次得分 | 较上次 | 目标 |\n|------|---------|--------|------|\n| 产品知识 | 82/100 | ↑5 | 80+ |\n| 异议处理 | 71/100 | ↓3 | 75+ |\n| 促成技巧 | 85/100 | ↑8 | 80+ |\n| 合规话术 | 95/100 | →0 | 100 |\n| 方案设计 | 72/100 | ↑12 | 75+ |\n\n### 🔥 本次表现亮点\n1. 养老规划方案逻辑清晰，能结合客户生命周期讲解\n2. 合规话术使用规范，犹豫期/退保说明完整\n\n### ⚠️ 需要加强\n1. 健康险异议处理：回应\"已有社保\"时过于被动，应主动算账\n2. 竞品对比：对中国平安主要产品线不够熟悉\n\n### 📅 明日训练重点\n- 产品：康健医疗保险（健康告知流程）\n- 场景：竞品替换（平安福 vs. XX福）\n- 时长：30分钟情景对练 + 10分钟快问快答\n```\n\n---\n\n## Workflow / 标准工作流程\n\n> **⚠️ 重要提示**：以下工作流展示的是**培训场景的教学参考**。所有销售话术和异议处理内容均为培训素材，实际销售行为须遵循《保险法》及相关监管规定，经持牌保险专业人士审核。\n\n### Mode 1: Quick Start (已知产品 + 快速训练)\n\n```\nUser: \"帮我准备明天拜访客户B的训练，他是私企老板，对健康险感兴趣\"\n │\n ▼\n[Step 1] 获取代理人信息 → 张明，L2，弱项：健康险异议处理\n[Step 2] 识别拜访产品 → 康健医疗保险（目标：替换平安福）\n[Step 3] 生成训练计划 → 午间30分钟：健康险异议处理对练\n[Step 4] 开始陪练 → 情景对练：私企业主健康险需求挖掘\n[Step 5] 实时反馈 → 异议处理评分：71/100，给出改进建议\n[Step 6] 报告输出 → 训练报告 + 明日拜访话术优化建议\n```\n\n### Mode 2: Product Document Upload (上传产品文档)\n\n```\nUser: [上传 XX保险公司福享人生终身寿险 产品手册 PDF]\n │\n ▼\n[Step 1] 解析文档 → 提取产品结构、条款、卖点\n[Step 2] 生成产品画像 → Structured JSON Profile\n[Step 3] 生成问题库 → 65+道题目（5难度×8类别）\n[Step 4] 生成参考题库 → 作为AI对话上下文，**不持久存储**\n[Step 5] 等待选择 → \"请选择训练模式：快问快答 / 情景对练 / 案例研讨\"\n```\n\n### Mode 3: Full Agent Assessment (全面能力评估)\n\n```\nUser: \"帮我评估代理人李华的综合能力，她入职8个月，主要卖重疾险\"\n │\n ▼\n[Step 1] 建立代理人档案 → L1入门级，8个月，重疾险方向\n[Step 2] 产品文档上传 → 重疾险产品手册\n[Step 3] 综合考核 → 30题产品知识 + 5个情景对练\n[Step 4] 生成能力雷达图 → 6维度能力可视化\n[Step 5] 制定成长路径 → 90天训练计划\n```\n\n---\n\n## Input / Output Specifications / 输入输出规范\n\n### Input\n\n| Input Type | Description | Example |\n|------------|-------------|---------|\n| 代理人档案 | JSON/文本描述 | 姓名、级别、工龄、业绩、弱项 |\n| 产品文档 | PDF/Word/TXT/图片 | 保险产品手册、条款、计划书 |\n| 当日行程 | 文本/日历 | 09:00晨会 / 10:30拜访客户A |\n| 训练指令 | 自然语言 | \"帮我准备健康险的陪练\" |\n| 客户信息 | 文本描述 | \"45岁私企老板，年收入200万\" |\n\n### Output\n\n| Output Type | Description |\n|-------------|-------------|\n| 产品画像JSON | 结构化产品信息 |\n| 问题库 | 65+道分类分级题目 |\n| 训练计划 | 分钟级个性化日程 |\n| 陪练对话 | 实时AI角色扮演 |\n| 评估报告 | 评分 + 改进建议 + 雷达图 |\n| 成长路径 | 30/60/90天训练建议 |\n\n---\n\n## Integration Notes / 集成说明\n\n**Data privacy:**\n- Treat all agent and client data as sensitive: the user decides where it is stored.\n This skill itself stores nothing; keep PII inside your own controlled systems\n- No sensitive PII should be included in training documents\n- Comply with China CBIRC insurance sales compliance regulations\n\n**Lianxi with other Skills:**\n- `insurance-bidding-pro`: Use product analysis for bidding scenarios\n- `insurance-private-domain-ops`: Link training completion to customer follow-up\n- `insurance-claims-intelligence`: Train agents on claim processes for better client communication\n\n---\n\n## 保存与外发前确认 / Save & Send Confirmation（2026-09-24 新增）\n\n1. 训练报告、话术稿、题库先在对话中完整展示，供用户逐段预览；\n2. 用户明确确认后再写入文件或对外发送；未经确认不落盘、不发送；\n3. 落盘时一并记录：生成时间、适用的产品条款版本、审核人。\n\n---\n\n## 变更记录 / Changelog\n\n| 版本 | 日期 | 变更摘要 |\n|------|------|---------|\n| 5.2.5 | 2026-09-24 | 新增文档解析实操示例、能力等级评定示例、红线词扫雷与续期服务演练两种模式；环境变化表新增“对应培训模块/可信度标注”两列并补充报行合一、销售可回溯两条；题库与模式表新增维度列；修复 3 处未闭合代码块；修正能力与代码示例不一致的声明 |\n| 5.2.4 | 2026-08-29 | 补充 2026 年预定利率与营销合规环境变化 |\n\n---\n\n## Disclaimer / 免责声明\n\n> ⚠️ **Training is advisory only.** This skill provides coaching materials, question banks,\n> and simulation training for insurance agent development. All final sales advice,\n> compliance decisions, and product recommendations must be reviewed by licensed\n> insurance professionals and comply with CBIRC regulations. Model answers represent\n> reference best practices, not guaranteed outcomes.\n\nFile v5.2.5:_meta.json\n\n{\n \"ownerId\": \"kn74e704j3ygjcygnpf02rdvd185js13\",\n \"slug\": \"insurance-agent-trainer\",\n \"version\": \"5.2.5\",\n \"publishedAt\": 1790227032726\n}\n\nFile v5.2.5:skill-card.md\n\n## Description:\n\nAI-powered insurance agent training coach that parses product documents conceptually, generates question banks, assesses agent skill levels, schedules personalized daily training, and supports interactive role-play sessions.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[gechengling](https://clawhub.ai/user/gechengling)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nInsurance training teams, agency managers, and agents use this skill to prepare coaching materials, question banks, role-play drills, training plans, and advisory assessment reports for insurance-agent development. Outputs require human review before real sales, compliance, or product-recommendation use.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Training scripts or regulatory notes could be mistaken for current licensed insurance, legal, or financial advice.\n\nMitigation: Treat outputs as training examples only and require licensed compliance review before using generated material in real sales contexts.\n\nRisk: Agent or customer details entered during training could include sensitive personal information.\n\nMitigation: Avoid entering real customer PII, use desensitized descriptions, and keep any training records limited to summary metrics unless proper consent and controls are in place.\n\nRisk: Insurance rules, product terms, and sales-compliance requirements may change after the skill content was written.\n\nMitigation: Verify current insurance regulations and product terms against authoritative sources before applying any training output.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/gechengling/skills/insurance-agent-trainer)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, code, configuration, guidance]\n\n**Output Format:** [Markdown guidance with structured JSON examples and illustrative pseudocode.]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Advisory educational outputs; no executable package, persistent storage, network calls, or credential collection are bundled.]\n\n## Skill Version(s):\n\n5.2.5 (source: frontmatter and server release metadata)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v5.2.4: 3 files, 14766 bytes\n\nFiles: skill-card.md (2567b), SKILL.md (32362b), _meta.json (142b)\n\nFile v5.2.4:SKILL.md\n\n---\nname: Insurance Agent Intelligent Trainer\ndescription: >\n AI-powered insurance agent training coach — auto-parses product docs, generates question banks,\n assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training\n based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers\n predinig rate cut (3.0%) impact on sales scripts, new health insurance regulation, PIPL-compliant\n customer communication, and benchmark against AIA, Ping An, and Alibaba Cloud training systems.\n Keywords: 智能陪练, 代理人培训, 保险训练, 产品陪练, 话术训练, 角色扮演, 技能评估, AIA培训, 平安培训, 代理人展业, 保险销售, 客户面谈, 异议处理, 保险培训系统, 学习路径.\nslug: insurance-agent-trainer\nversion: 5.2.4\nallowed-tools: []\ncapabilities:\n - educational-reference\n - advisory-only\n - requires-human-review\n - no-executable-code\n---\n\n# Insurance Agent Intelligent Trainer / 保险代理人智能陪练系统\n\n> **⚠️ SECURITY NOTICE / 安全声明**\n> - **Type:** Educational reference / analytical framework ONLY\n> - **No executable code, scripts, or binaries are included in this skill**\n> - **No persistent storage, network calls, background execution, or credential collection**\n> - **All outputs are for reference only and require human review before real-world application**\n> - **This skill does NOT provide financial, legal, or insurance advice**\n> - **Users must exercise their own judgment and consult qualified professionals**\n>\n> **⚠️ 数据安全警告**\n> - 本技能仅提供保险代理人的培训辅导参考框架，**不执行任何代码或脚本**\n> - 所有文档解析、日程分析、画像评估的描述均为**教学参考框架**，**不包含实际的OCR或PDF解析引擎**\n> - 不会自动访问、存储或处理用户的任何培训数据或个人信息\n> - 培训计划和话术建议需结合用户实际业务场景调整，**不能替代专业培训师**\n> - **销售话术和异议处理仅为培训参考，实际使用须遵守《保险法》及相关监管规定，不得以AI输出替代合规审核**\n\n\n\n> **English:** AI-powered insurance agent coaching system — parses product documents, generates\n> personalized question banks, assesses agent competency levels, schedules daily training based on\n> client visits, and runs interactive role-play drills. Benchmarked against AIA, Ping An, and\n> Alibaba Cloud insurance training systems.\n>\n> **中文:** 保险代理人智能陪练系统——解析产品文档、自动生成问题库、评估代理人能力等级、\n> 结合当日客户拜访行程安排个性化训练、进行情景对练。对标友邦保险、平安保险、阿里云智能陪练水平。\n\n---\n\n## Trigger Keywords / 触发关键词\n\n**⚠️ 精确触发规则**：仅当用户明确提到保险代理人培训/陪练相关需求时激活。日常对话中提及\"培训\"、\"训练\"、\"coaching\"、\"agent training\"等通用词汇时**不会自动触发**。\n\n**用户确认规则**：当用户输入匹配以下关键词时，必须先确认用户意图：\n- \"您需要保险代理人陪练/培训服务吗？\"\n- 仅在用户明确确认后，才进入陪练模式\n\n激活关键词（需用户确认后生效）：\n\n- 保险陪练 / 产品陪练 / 智能陪练 / 代理人训练\n- 代理人培训 / 新人培训 / 保险话术训练\n- 产品演练 / 客户异议处理 / 保险销售训练\n- insurance agent training / insurance coaching / insurance product drill\n\n---\n\n## Core System Architecture / 核心系统架构\n\n### 0. 2025-2026 代理人销售环境最新变化（截至2026-08）\n\n| 变化 | 内容 | 话术调整建议 |\n|------|------|------------|\n| **预定利率降至3.0%** | 2024年9月后所有新产品执行 | 强调\"锁定3.0%长期确定收益\"，对比银行理财波动性 |\n| **分红险主导市场** | 分红险、万能险替代传统高利率产品 | 学会讲\"浮动收益+保底保障\"的双重价值 |\n| **健康险新规上线** | 2025年商业健康险管理办法修订 | 健康告知流程需更规范，禁止误导性说明 |\n| **代理人资格考试升级** | 2025年加入AI伦理、数字化服务模块 | 新人需补充数字化能力培训 |\n| **企微客户触达合规** | AI外呼需标注身份，营销需客户授权 | 培训合规营销话术，避免违规外呼 |\n| **预定利率进一步下调至2.0%** | 2026年监管引导普通型人身险预定利率上限降至2.0%，分红/万能演示利率同步压降 | 话术从\"锁定3.0%\"转为\"锁定2.0%长期确定+浮动分红对冲通胀\" |\n| **营销宣传合规强化** | 2026年整治\"炒停售\"\"夸大收益\"，自媒体/直播带货纳入监管 | 培训合规表达，禁用绝对化收益承诺与演示红线 |\n| **养老金融与税优扩容** | 个人养老金、商业养老金试点扩围，税优额度可期上调 | 强化养老规划与税优测算话术，绑定家庭现金流诊断 |\n\n\n\n```\n┌─────────────────────────────────────────────────────────────────┐\n│ Insurance Agent Intelligent Trainer │\n├─────────────────────────────────────────────────────────────────┤\n│ ┌──────────────┐ ┌──────────────┐ ┌──────────────────────┐ │\n│ │ Product Doc │ │ Agent Profile│ │ Daily Schedule/Routes│ │\n│ │ Parser │ │ Engine │ │ Integration │ │\n│ │ (PDF/Word/ │ │ (Skill Level │ │ (Today's Visits & │ │\n│ │ Images) │ │ Assessment) │ │ Client Profiles) │ │\n│ └──────┬───────┘ └──────┬───────┘ └──────────┬───────────┘ │\n│ │ │ │ │\n│ ▼ ▼ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Question Bank Generation Engine │ │\n│ │ Product Knowledge │ Objection Handling │ Case Analysis │ │\n│ │ [5 difficulty tiers × 3 categories = 15 question types] │ │\n│ └──────────────────────────┬───────────────────────────────┘ │\n│ │ │\n│ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Personalized Training Scheduler │ │\n│ │ [Skill Level + Schedule + Product Priority = Daily Plan]│ │\n│ └──────────────────────────┬───────────────────────────────┘ │\n│ │ │\n│ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Interactive Training Engine │ │\n│ │ Role-play │ Real-time Feedback │ Progress Tracking │ │\n│ └──────────────────────────────────────────────────────────┘ │\n└─────────────────────────────────────────────────────────────────┘\n```\n\n---\n\n## Core Capabilities / 核心能力\n\n### 1. Product Document Parser / 产品文档解析引擎（教学演示）\n\n> **⚠️ 教学演示**：以下展示产品文档解析的**概念性教学方法论**，仅说明AI可如何辅助理解产品结构。**本技能不执行任何实际的PDF解析、OCR识别或文档提取操作。** 所有\"解析流程\"均为逻辑示意，实际应用需由具体的工程实现完成。\n\n**Supported formats (conceptual):** PDF, Word (.docx), scanned images (with OCR), plain text\n\n**Conceptual parsing pipeline (for reference):**\n\n```\nDocument Upload\n │\n ▼\n[Format Detection] → PDF / Word / Image / Text\n │\n ▼\n[Text Extraction] → Raw text content\n │\n ▼\n[Structure Analysis]\n ├─ Product name, type, target customers\n ├─ Coverage scope (death, medical, annuity, critical illness, etc.)\n ├─ Premium levels & payment periods\n ├─ Policy terms & exclusions\n ├─ Sales pitch key points\n ├─ Competitive advantages vs. similar products\n └─ Compliance notes & regulatory requirements\n │\n ▼\n[Structured Product Profile] → Ready for question generation\n```\n\n**Output: Structured Product Profile JSON**\n\n```json\n{\n \"product_name\": \"XX福享人生终身寿险(万能型)\",\n \"product_type\": \"whole-life insurance with universal account\",\n \"insurer\": \"国联人寿\",\n \"target_customers\": [\"30-50岁中高收入人群\", \"有财富传承需求\"],\n \"coverage\": {\n \"death_benefit\": \"100%-160%账户价值\",\n \"annuity_option\": \"60岁起可转换为年金\",\n \"waiver\": \"可选投保人保费豁免\"\n },\n \"premium\": {\n \"min_annual\": 12000,\n \"payment_periods\": [\"3年\", \"5年\", \"10年\", \"20年\"],\n \"min_coverage_years\": \"终身\"\n },\n \"key_selling_points\": [\n \"复利增值，万能账户历史结算利率4.5%-5.2%\",\n \"灵活追加，额外资金可随时进入万能账户\",\n \"身故保障与财富传承双重功能\"\n ],\n \"competitive_edges\": [\"结算利率优于同类竞品\", \"追加无上限\"],\n \"exclusions\": [\"投保人对被保险人的故意伤害\", \"2年内自杀(无民事行为能力人除外)\"],\n \"compliance_notes\": [\"需双录(录音录像)\", \"犹豫期15天\", \"等待期90天\"],\n \"difficulty_tags\": [\"新人友好\", \"需强化健康告知\", \"财务规划综合能力\"]\n}\n\n**示例产品画像 2（重大疾病保险）：**\n```json\n{\n \"product_name\": \"XX康健终身重疾险(2026版)\",\n \"product_type\": \"critical illness insurance\",\n \"insurer\": \"国联人寿\",\n \"target_customers\": [\"28-50岁家庭经济支柱\", \"有重疾保障缺口人群\"],\n \"coverage\": {\n \"ci_types\": \"120种重疾+20种中症+40种轻症\",\n \"multiple_payout\": \"重疾1次+中症2次+轻症3次，累计最高260%保额\",\n \"death_benefit\": \"身故赔已交保费或现金价值较大者\"\n },\n \"premium\": {\n \"sample\": \"30岁男，50万保额，30年缴，年缴约 6800 元\",\n \"payment_periods\": [\"10年\",\"20年\",\"30年\"]\n },\n \"key_selling_points\": [\n \"重疾+中症+轻症三重递进保障\",\n \"轻中症豁免后续保费\",\n \"可附加恶性肿瘤二次赔付\"\n ],\n \"exclusions\": [\"投保前已患重疾\", \"遗传性疾病（条款约定）\", \"等待期内出险\"],\n \"compliance_notes\": [\"重疾定义以监管规范为准\", \"需明确告知等待期90-180天\", \"如实健康告知义务\"],\n \"difficulty_tags\": [\"健康告知敏感\", \"条款专业度高\", \"需结合医疗知识\"]\n}\n```\n```\n\n---\n\n### 2. Agent Profile & Skill Assessment / 代理人画像与能力评估\n\n> **⚠️ 数据处理提醒**：以下代理人画像和日程数据为**演示示例**。实际使用时，用户应自行管理代理人数据的收集和存储，确保符合《个人信息保护法》及保险行业合规要求。请勿输入真实客户PII信息。\n\n**Three skill tiers:**\n\n| Tier | Level | Description | Training Focus | 建议训练时长/周 |\n|------|-------|-------------|----------------|----------------|\n| 🌱 **L1 - 入门级** | Beginner | < 1 year experience, struggles with product details and objection handling | Foundation: product knowledge, basic sales scripts, simple objection responses | 5-8 小时（晨会快练+情景对练） |\n| ⚡ **L2 - 进阶级** | Intermediate | 1-3 years, solid product knowledge but inconsistent closing rate | Application: complex scenarios, multi-product combination, competitive replacement, high-net-worth clients | 3-5 小时（聚焦弱项情景对练） |\n| 🎯 **L3 - 专家级** | Advanced | 3+ years, high performance, needs strategy for complex cases | Mastery: enterprise/group clients, tax planning, estate planning, competitive stealing, mentoring skills | 2-3 小时（策略复盘+带教新人） |\n\n**Profile structure:**\n\n```json\n{\n \"agent_id\": \"AG20240001\",\n \"name\": \"张明\",\n \"level\": \"L2\",\n \"level_label\": \"进阶级\",\n \"tenure_years\": 2.5,\n \"certifications\": [\"保险代理人资格证\", \"健康险销售资质\"],\n \"performance\": {\n \"monthly_premium_target\": 50000,\n \"monthly_premium_actual\": 42000,\n \"closing_rate\": 0.32,\n \"avg_policy_size\": 18500,\n \"new_customer_rate\": 0.45\n },\n \"product_mastery\": {\n \"term_life\": 0.85,\n \"whole_life\": 0.72,\n \"critical_illness\": 0.58,\n \"medical_insurance\": 0.80,\n \"annuity\": 0.45,\n \"investment_linked\": 0.38\n },\n \"weak_points\": [\n \"健康险异议处理不够熟练\",\n \"不了解高端客户的税务筹划需求\",\n \"组合产品销售话术单一\"\n ],\n \"strong_points\": [\n \"老客户维护能力强\",\n \"缘故市场开拓优秀\"\n ],\n \"daily_schedule\": [\n {\"time\": \"09:00-10:00\", \"activity\": \"晨会\", \"location\": \"营业部\"},\n {\"time\": \"10:30-12:00\", \"activity\": \"拜访客户A（国企中层，有养老需求）\", \"location\": \"客户公司\"},\n {\"time\": \"14:00-15:30\", \"activity\": \"拜访客户B（私企业主，健康险需求）\", \"location\": \"客户公司\"},\n {\"time\": \"16:00-17:30\", \"activity\": \"缘故客户C（教育金规划）\", \"location\": \"咖啡厅\"}\n ]\n}\n\n**示例画像 2（L3 专家级）：**\n```json\n{\n \"agent_id\": \"AG20230088\",\n \"name\": \"李华\",\n \"level\": \"L3\",\n \"level_label\": \"专家级\",\n \"tenure_years\": 6,\n \"certifications\": [\"保险代理人资格证\", \"CFP国际金融理财师\", \"私人银行家\"],\n \"performance\": {\n \"monthly_premium_target\": 200000,\n \"monthly_premium_actual\": 235000,\n \"closing_rate\": 0.48,\n \"avg_policy_size\": 86000,\n \"new_customer_rate\": 0.62\n },\n \"product_mastery\": {\n \"term_life\": 0.95, \"whole_life\": 0.92, \"critical_illness\": 0.90,\n \"medical_insurance\": 0.93, \"annuity\": 0.88, \"investment_linked\": 0.82\n },\n \"weak_points\": [\"家族信托等复杂传承架构经验不足\", \"跨境税务筹划需外部专家协同\"],\n \"strong_points\": [\"高净值客户经营\", \"企业团险开拓\", \"复杂方案设计\"],\n \"coaching_focus\": [\"传承架构进阶\", \"监管合规红线强化\", \"带教新人方法论\"]\n}\n```\n```\n\n---\n\n### 3. Question Bank Generation / 问题库自动生成（教学模板）\n\n> **⚠️ 教学演示**：以下问题库和话术为**培训场景的教学参考模板**，展示如何结构化设计代理人训练内容。所有涉及销售话术、竞品对比、异议处理的内容均为**培训素材**，实际销售行为须遵循《保险法》及相关监管规定，并经持牌保险专业人士审核后方可执行。\n\n**Generated from product profile + agent level + training objectives**\n\n#### Question Types (15 categories across 3 dimensions)\n\n**By Category:**\n\n| Category | Description | Example | 考核重点 |\n|----------|-------------|---------|---------|\n| **产品知识** | Product features, terms, coverage | \"XX福的等待期是多久？\" | 条款准确性、关键利益点无误 |\n| **客户画像** | Target customer identification | \"什么样的客户适合购买这款产品？\" | 需求诊断与匹配逻辑 |\n| **异议处理** | Objection handling scripts | \"客户说'我已经有社保了，不需要商业保险'，如何回应？\" | 共情+数据化反驳能力 |\n| **案例分析** | Real case discussion | \"40岁国企中层，年薪50万，如何用这款产品做养老规划？\" | 方案完整性与定制化 |\n| **合规话术** | Compliance-approved scripts | \"如何向客户解释犹豫期和退保损失？\" | 红线词零触发 |\n| **竞品对比** | vs. competitors | \"相比平安福，这款产品的核心优势是什么？\" | 客观不贬损竞品 |\n| **促成话术** | Closing techniques | \"客户表现出购买意向，如何自然促成？\" | 时机把握自然度 |\n| **交叉销售** | Multi-product combination | \"如何将主险与医疗险组合销售？\" | 保障缺口覆盖度 |\n| **养老规划** | 养老现金流与替代率测算 | \"客户55岁期望退休月领8000，如何测算缺口？\" | 测算逻辑与工具使用 |\n| **税优保险** | 个人养老金/税优健康险政策应用 | \"年缴1.2万养老金，节税多少？\" | 政策准确、不夸大节税 |\n\n**By Difficulty (5 tiers):**\n\n| Level | Target Audience | Question Complexity | 建议题量/次 |\n|--------|----------------|---------------------|------------|\n| ⭐ 基础 | L1新人 | 单一产品，单一问题，直接答案 | 10-15 题 |\n| ⭐⭐ 入门 | L1-L2 | 单一产品，1-2个知识点，需要解释 | 15-20 题 |\n| ⭐⭐⭐ 进阶 | L2 | 单一产品，3-5个知识点，需组合分析 | 20-30 题 |\n| ⭐⭐⭐⭐ 高阶 | L2-L3 | 多产品组合，竞争替换，高净值客户 | 25-35 题 |\n| ⭐⭐⭐⭐⭐ 专家 | L3 | 综合方案，税务筹划，财富传承 | 30-40 题 |\n\n#### Question Bank Generation Prompt:\n\n```\nBased on the product profile provided, generate a question bank with:\n\n1. For each difficulty tier (基础/入门/进阶/高阶/专家):\n - 5 multiple choice questions (产品知识)\n - 3 case analysis questions\n - 3 objection handling scenarios\n - 2 competitive comparison questions\n - 1 closing technique exercise\n\n2. Total: 65+ questions per product\n\n3. For each question, provide:\n - Question text\n - Difficulty level (1-5)\n - Category (产品知识/异议处理/案例分析/竞品对比/促成话术)\n - Ideal answer / model response\n - Evaluation criteria (excellent/good/needs-improvement)\n - Coaching tips for the trainer\n```\n\n**示例生成题目（养老规划类别 / ⭐⭐⭐ 进阶）：**\n- **题目**：客户 55 岁，当前社保养老金预计月领 3500 元，期望退休后月生活支出 8000 元，如何测算商业养老金缺口？\n- **参考答案**：缺口 = (8000 - 3500) × 12 × 退休年限（按 25 年计）≈ 135 万；结合预期投资收益率反推年缴/趸交金额，并叠加通胀与医疗支出弹性。\n- **评分**：优秀（准确测算+工具使用）/ 良好（逻辑正确但忽略通胀）/ 待改进（未考虑长寿风险）。\n- **教练提示**：引导代理人用\"替代率\"概念切入，避免直接推销产品。\n\n---\n\n### 4. Personalized Training Scheduler / 个性化训练调度引擎（方法论演示）\n\n> **⚠️ 教学演示**：以下调度算法、代理人画像及日程数据均为**教学方法论的概念性展示**。**本技能不实际采集、存储或处理任何代理人或客户数据**。所有姓名、日程、业绩数据均为虚构示例，仅用于说明逻辑框架。\n\n**Input factors:**\n\n```\nAgent Profile (Level + Weak Points)\n +\nToday's Client Schedule (Who → What need → What product)\n +\nProduct Priority Matrix\n =\nPersonalized Daily Training Plan\n```\n\n**Scheduling Algorithm:**\n\n```python\ndef generate_daily_training_plan(agent_profile, daily_schedule, products):\n \"\"\"\n Generate personalized training plan for the day.\n \"\"\"\n # Step 1: Identify today's client visit products\n today_products = extract_products_from_schedule(daily_schedule)\n \n # Step 2: Get agent's weakness areas for these products\n weakness_map = get_weakness_for_products(\n agent_profile.weak_points, \n today_products\n )\n \n # Step 3: Calculate training time available\n available_minutes = calculate_available_training_time(daily_schedule)\n \n # Step 4: Prioritize by impact × weakness × product value\n training_queue = prioritize_training(\n weakness_map,\n today_products,\n agent_profile.level,\n time_constraint=available_minutes\n )\n \n # Step 5: Generate session plan\n sessions = split_into_sessions(training_queue, available_minutes)\n \n return {\n \"date\": today,\n \"agent\": agent_profile.name,\n \"total_minutes\": available_minutes,\n \"sessions\": sessions,\n \"focus_products\": today_products,\n \"key_objectives\": get_key_objectives(training_queue)\n }\n```\n\n**Example Daily Training Plan:**\n\n```json\n{\n \"date\": \"2026-05-05\",\n \"agent\": \"张明\",\n \"level\": \"L2\",\n \"total_minutes\": 90,\n \"sessions\": [\n {\n \"time\": \"08:00-08:20\",\n \"duration\": 20,\n \"type\": \"晨间快练\",\n \"mode\": \"快问快答\",\n \"focus\": \"年金险产品知识（高频问题5题）\",\n \"product\": \"福享人生终身寿险\",\n \"objective\": \"巩固年金转换权的计算逻辑\"\n },\n {\n \"time\": \"12:30-13:00\",\n \"duration\": 30,\n \"type\": \"午间强化\",\n \"mode\": \"情景对练\",\n \"focus\": \"健康险异议处理\",\n \"scenario\": \"客户：\"我有社保，不需要商业医疗险\"\",\n \"product\": \"康健医疗保险\",\n \"level\": \"⭐⭐⭐ 进阶\",\n \"coaching_tips\": \"引导客户认识到社保报销比例上限，用自费药比例对比引发需求\"\n },\n {\n \"time\": \"17:30-18:30\",\n \"duration\": 40,\n \"type\": \"晚间复盘\",\n \"mode\": \"案例分析 + 角色扮演\",\n \"focus\": \"私企业主综合保障方案\",\n \"scenario\": \"45岁私企老板，年收入200万，已有多份保单，如何做加保方案？\",\n \"products\": [\"终身寿险+万能账户\", \"高端医疗\", \"企业财产险\"],\n \"level\": \"⭐⭐⭐⭐ 高阶\",\n \"model_response_guide\": \"从家庭资产与企业资产隔离角度切入，引出终身寿险的债务隔离和传承功能\"\n }\n ],\n \"key_metrics_to_track\": [\n \"异议处理响应时间（目标<30秒）\",\n \"产品知识点正确率（目标>85%）\",\n \"方案组合完整性（3单以上产品覆盖）\"\n ]\n}\n\n**示例计划 2（健康险异议攻坚日）：**\n```json\n{\n \"date\": \"2026-08-12\",\n \"agent\": \"李华\",\n \"level\": \"L3\",\n \"total_minutes\": 60,\n \"sessions\": [\n {\n \"time\": \"13:00-13:20\",\n \"duration\": 20,\n \"type\": \"午间强化\",\n \"mode\": \"异议攻关\",\n \"focus\": \"健康险'已有社保'高频异议\",\n \"scenario\": \"客户：'我有医保，重疾险没必要'\",\n \"level\": \"⭐⭐⭐ 进阶\",\n \"coaching_tips\": \"用'医保目录外用药+收入补偿'双轴拆解，量化缺口而非否定客户\"\n },\n {\n \"time\": \"18:00-18:40\",\n \"duration\": 40,\n \"type\": \"晚间复盘\",\n \"mode\": \"案例研讨 + 角色扮演\",\n \"focus\": \"高净值客户重疾+医疗+寿险组合\",\n \"scenario\": \"50岁企业主，家庭年收入300万，已有多张保单如何查漏补缺？\",\n \"level\": \"⭐⭐⭐⭐ 高阶\",\n \"model_response_guide\": \"从企业资产与家庭资产隔离、重疾收入补偿、医疗高端资源三维度切入\"\n }\n ],\n \"key_metrics_to_track\": [\n \"异议处理响应时间（目标<25秒）\",\n \"方案组合维度（目标≥4个）\",\n \"合规红线触发（目标0次）\"\n ]\n}\n```\n```\n\n---\n\n### 5. Interactive Training Session / 智能陪练对话引擎\n\n**Session modes:**\n\n| Mode | Description | Duration | Best For | 适用场景 |\n|------|-------------|----------|----------|---------|\n| **快问快答** | Rapid-fire Q&A | 5-10 min | Pre-meeting warmup | 晨会热身、拜访前激活产品知识 |\n| **情景对练** | Role-play (client vs. agent) | 15-30 min | Skill practice | 健康险/养老险高频异议实战 |\n| **案例研讨** | Real case analysis | 20-40 min | Advanced agents | 高净值综合保障方案设计 |\n| **异议攻关** | Objection busting focus | 10-15 min | Weak point training | 单一弱项（如\"已有社保\"）专项突破 |\n| **综合考核** | Full simulation exam | 30-60 min | Level assessment | 晋升/季度能力认证 |\n| **直播带练** | 模拟自媒体/直播讲保险 | 15-25 min | 数字化展业 | 合规表达与镜头前讲产品 |\n\n**Real-time coaching during training:**\n\n```\nAgent Response\n │\n ▼\n[Natural Language Understanding] → Extract key claims, tone, strategy\n │\n ▼\n[Evaluation Engine]\n ├─ Product knowledge accuracy ✓/✗\n ├─ Objection handling effectiveness (1-5)\n ├─ Compliance adherence ✓/✗\n ├─ Closing attempt timing (good/early/late/missing)\n ├─ Client empathy signals ✓/✗\n └─ Product combination logic ✓/✗\n │\n ▼\n[Real-time Coaching Feedback]\n ├─ Immediate tip (if struggling): \"💡 提示：可以先问客户目前的保障缺口...\"\n ├─ Completion praise (if excellent): \"🌟 完美！您已经很好地识别了客户需求\"\n └─ Post-question summary: \"本轮得分 85/100。建议加强：竞品对比环节\"\n```\n\n**Training session flow:**\n\n```\n1. 导入 (5%) → 介绍训练目标和产品背景\n2. 暖场 (10%) → 快问快答热身，激活产品知识\n3. 主体 (60%) → 情景对练：客户角色扮演 + 实时点评\n4. 复盘 (20%) → AI给出详细反馈：优点/不足/改进建议\n5. 行动 (5%) → 下次拜访的具体行动计划\n```\n\n**示例对练对话片段（健康险异议攻关）：**\n```\nAI(客户): \"我单位福利好，重疾险真没必要买。\"\nAgent: \"您说的对，单位福利是重要保障。不过重疾理赔是'确诊即付'的一笔钱——您想过没有，万一需要长期康复，单位会不会照发全额工资？\"\nAI(客户): \"那倒不会，病假工资大概只发底薪……\"\nAgent: \"这就是缺口。重疾险补的正是'收入中断+自费药'这两块。我们按您月支出算一下具体差额？\"\n[实时点评] ✅ 共情到位；✅ 用'收入补偿'替代'恐吓式'话术；⚠️ 下一步应主动给出测算而非直接推产品。\n```\n\n---\n\n### 6. Effect Assessment & Progress Tracking / 效果评估与进度追踪\n\n**Metrics tracked per session:**\n\n| Metric | Definition | Target | 评估方式 |\n|--------|------------|--------|---------|\n| **产品知识得分** | 知识点正确率 | L1: ≥70%, L2: ≥80%, L3: ≥90% | 自动判分题库 + 人工抽检 |\n| **异议处理时效** | 从异议提出到满意回答的时间 | < 30秒 | 对话时间戳测算 |\n| **促成成功率** | 能否自然引入促成信号 | ≥ 1次有效尝试 | 教练引擎标记 + 主管复核 |\n| **话术合规率** | 合规敏感词使用正确性 | 100% | 合规词库实时监测 |\n| **方案完整性** | 保障覆盖广度 | ≥ 3个维度 | 结构化评分表 |\n| **合规红线触发率** | 触碰禁语/误导表述次数 | 0 次 | 实时拦截 + 事后复盘 |\n\n**Progress report structure:**\n\n```markdown\n## 📊 代理人张明 训练报告 - 2026-05-05\n\n### 综合得分: ⭐⭐⭐⭐ (78/100)\n\n| 维度 | 本次得分 | 较上次 | 目标 |\n|------|---------|--------|------|\n| 产品知识 | 82/100 | ↑5 | 80+ |\n| 异议处理 | 71/100 | ↓3 | 75+ |\n| 促成技巧 | 85/100 | ↑8 | 80+ |\n| 合规话术 | 95/100 | →0 | 100 |\n| 方案设计 | 72/100 | ↑12 | 75+ |\n\n### 🔥 本次表现亮点\n1. 养老规划方案逻辑清晰，能结合客户生命周期讲解\n2. 合规话术使用规范，犹豫期/退保说明完整\n\n### ⚠️ 需要加强\n1. 健康险异议处理：回应\"已有社保\"时过于被动，应主动算账\n2. 竞品对比：对中国平安主要产品线不够熟悉\n\n### 📅 明日训练重点\n- 产品：康健医疗保险（健康告知流程）\n- 场景：竞品替换（平安福 vs. XX福）\n- 时长：30分钟情景对练 + 10分钟快问快答\n```\n\n---\n\n## Workflow / 标准工作流程\n\n> **⚠️ 重要提示**：以下工作流展示的是**培训场景的教学参考**。所有销售话术和异议处理内容均为培训素材，实际销售行为须遵循《保险法》及相关监管规定，经持牌保险专业人士审核。\n\n### Mode 1: Quick Start (已知产品 + 快速训练)\n\n```\nUser: \"帮我准备明天拜访客户B的训练，他是私企老板，对健康险感兴趣\"\n │\n ▼\n[Step 1] 获取代理人信息 → 张明，L2，弱项：健康险异议处理\n[Step 2] 识别拜访产品 → 康健医疗保险（目标：替换平安福）\n[Step 3] 生成训练计划 → 午间30分钟：健康险异议处理对练\n[Step 4] 开始陪练 → 情景对练：私企业主健康险需求挖掘\n[Step 5] 实时反馈 → 异议处理评分：71/100，给出改进建议\n[Step 6] 报告输出 → 训练报告 + 明日拜访话术优化建议\n```\n\n### Mode 2: Product Document Upload (上传产品文档)\n\n```\nUser: [上传 XX保险公司福享人生终身寿险 产品手册 PDF]\n │\n ▼\n[Step 1] 解析文档 → 提取产品结构、条款、卖点\n[Step 2] 生成产品画像 → Structured JSON Profile\n[Step 3] 生成问题库 → 65+道题目（5难度×8类别）\n[Step 4] 生成参考题库 → 作为AI对话上下文，**不持久存储**\n[Step 5] 等待选择 → \"请选择训练模式：快问快答 / 情景对练 / 案例研讨\"\n```\n\n### Mode 3: Full Agent Assessment (全面能力评估)\n\n```\nUser: \"帮我评估代理人李华的综合能力，她入职8个月，主要卖重疾险\"\n │\n ▼\n[Step 1] 建立代理人档案 → L1入门级，8个月，重疾险方向\n[Step 2] 产品文档上传 → 重疾险产品手册\n[Step 3] 综合考核 → 30题产品知识 + 5个情景对练\n[Step 4] 生成能力雷达图 → 6维度能力可视化\n[Step 5] 制定成长路径 → 90天训练计划\n```\n\n---\n\n## Input / Output Specifications / 输入输出规范\n\n### Input\n\n| Input Type | Description | Example |\n|------------|-------------|---------|\n| 代理人档案 | JSON/文本描述 | 姓名、级别、工龄、业绩、弱项 |\n| 产品文档 | PDF/Word/TXT/图片 | 保险产品手册、条款、计划书 |\n| 当日行程 | 文本/日历 | 09:00晨会 / 10:30拜访客户A |\n| 训练指令 | 自然语言 | \"帮我准备健康险的陪练\" |\n| 客户信息 | 文本描述 | \"45岁私企老板，年收入200万\" |\n\n### Output\n\n| Output Type | Description |\n|-------------|-------------|\n| 产品画像JSON | 结构化产品信息 |\n| 问题库 | 65+道分类分级题目 |\n| 训练计划 | 分钟级个性化日程 |\n| 陪练对话 | 实时AI角色扮演 |\n| 评估报告 | 评分 + 改进建议 + 雷达图 |\n| 成长路径 | 30/60/90天训练建议 |\n\n---\n\n## Integration Notes / 集成说明\n\n**Data privacy:**\n- All agent and client data remains local / within the company's system\n- No sensitive PII should be included in training documents\n- Comply with China CBIRC insurance sales compliance regulations\n\n**Lianxi with other Skills:**\n- `insurance-bidding-pro`: Use product analysis for bidding scenarios\n- `insurance-private-domain-ops`: Link training completion to customer follow-up\n- `insurance-claims-intelligence`: Train agents on claim processes for better client communication\n\n---\n\n## Disclaimer / 免责声明\n\n> ⚠️ **Training is advisory only.** This skill provides coaching materials, question banks,\n> and simulation training for insurance agent development. All final sales advice,\n> compliance decisions, and product recommendations must be reviewed by licensed\n> insurance professionals and comply with CBIRC regulations. Model answers represent\n> reference best practices, not guaranteed outcomes.\n\nFile v5.2.4:_meta.json\n\n{\n \"ownerId\": \"kn74e704j3ygjcygnpf02rdvd185js13\",\n \"slug\": \"insurance-agent-trainer\",\n \"version\": \"5.2.4\",\n \"publishedAt\": 1788012356627\n}\n\nFile v5.2.4:skill-card.md\n\n## Description:\n\nAI-powered insurance agent training coach that helps structure product profiles, question banks, skill assessments, daily training plans, and role-play sessions for insurance agent development.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[gechengling](https://clawhub.ai/user/gechengling)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nExternal insurance trainers, sales managers, and agents use this skill to generate coaching materials, practice scenarios, question banks, assessment reports, and 30/60/90-day development plans for insurance sales training. Outputs are advisory training references and require qualified insurance or compliance review before real-world use.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Users may treat generated sales scripts, product recommendations, or objection-handling examples as financial, legal, insurance, or compliance advice.\n\nMitigation: Require licensed insurance or compliance staff to review final sales advice, product recommendations, and client-facing scripts before use.\n\nRisk: Training prompts may include real customer PII, confidential policy files, or regulated business data.\n\nMitigation: Do not paste real customer PII or confidential policy data unless the operating environment and company policies explicitly allow it.\n\nRisk: The skill describes product-document parsing and training workflows that could be mistaken for an operational compliance or parser system.\n\nMitigation: Treat parsing, scoring, scheduling, and coaching examples as educational frameworks; deploy separate reviewed systems for production document processing or compliance enforcement.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/gechengling/skills/insurance-agent-trainer)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, configuration, guidance]\n\n**Output Format:** [Markdown with structured JSON examples, tables, training plans, dialogue scripts, and assessment reports]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Produces advisory insurance training content only; no executable code, tools, network calls, or persistent storage are included.]\n\n## Skill Version(s):\n\n5.2.4 (source: frontmatter and server release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v5.2.3: 3 files, 13010 bytes\n\nFiles: skill-card.md (2552b), SKILL.md (28653b), _meta.json (142b)\n\nFile v5.2.3:SKILL.md\n\n---\r\nname: Insurance Agent Intelligent Trainer\r\ndescription: >\r\n AI-powered insurance agent training coach — auto-parses product docs, generates question banks,\r\n assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training\r\n based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers\r\n predinig rate cut (3.0%) impact on sales scripts, new health insurance regulation, PIPL-compliant\r\n customer communication, and benchmark against AIA, Ping An, and Alibaba Cloud training systems.\r\n Keywords: 智能陪练, 代理人培训, 保险训练, 产品陪练, 话术训练, 角色扮演, 技能评估, AIA培训, 平安培训, 代理人展业, 保险销售, 客户面谈, 异议处理, 保险培训系统, 学习路径.\r\nslug: insurance-agent-trainer\r\nversion: 5.2.3\r\nallowed-tools:\r\n - python-scripts\r\n - file-processing\r\ncapabilities:\r\n - educational-reference\r\n - analytical-framework\r\n - requires-human-review\r\n - training-content-generation\r\n---\r\n\r\n# Insurance Agent Intelligent Trainer / 保险代理人智能陪练系统\r\n\r\n> **⚠️ 能力声明 / Capability Notice**\r\n> - **Type:** Training framework with reference scripts — provides educational content generation and assessment templates\r\n> - **Bundled scripts:** Product parser, question generator, and training scheduler (Python) for local educational use\r\n> - **No persistent storage, network calls, background execution, or credential collection**\r\n> - **All outputs are for reference only and require human review before real-world application**\r\n> - **This skill does NOT provide financial, legal, or insurance advice**\r\n> - **Users must exercise their own judgment and consult qualified professionals**\r\n>\r\n> **⚠️ 使用声明**\r\n> - 本技能提供保险代理人的培训辅导框架，附带产品解析、题库生成、训练调度等参考脚本\r\n> - 所有脚本仅供本地教育参考使用，**不涉及外部API调用、数据采集或网络通信**\r\n> - 不会自动访问、存储或处理用户的任何培训数据或个人信息\r\n> - 培训计划和话术建议需结合用户实际业务场景调整，**不能替代专业培训师**\r\n> - **销售话术和异议处理仅为培训参考，实际使用须遵守《保险法》及相关监管规定，不得以AI输出替代合规审核**\r\n\r\n### 保险监管最新动态 [2026-07-18更新]\r\n\r\n| 动态类型 | 内容摘要 | 发布时间 | 影响范围 |\r\n|---------|---------|---------|---------|\r\n| 监管施行 | 分红险演示利率上限由3.9%下调至3.5%，各公司须在6月30日前完成变更备案或停售 | 2026-07-01 | 代理人培训需更新分红险话术与利益演示 |\r\n| 监管施行 | 《保险产品适当性管理自律规范》正式施行，产品P1-P5五级管理 | 2026-07-01 | 代理人需按资质等级销售对应产品 |\r\n| 监管施行 | 银保渠道65号文费用新规全面执行，全口径分项备案 | 2026-07-01 | 银保销售流程与费用管理全面规范 |\r\n| 监管发布 | 金融监管总局发布《银行业保险业AI安全开发应用指导意见》 | 2026-06-18 | 保险AI培训系统需符合合规要求 |\r\n| 草案发布 | 《金融业网络安全管理办法(征求意见稿)》发布，8章72条 | 2026-07-03 | 培训系统数据安全合规升级 |\r\n\r\n> **数据截止**: 2026-07-18 | 来源：国家金融监督管理总局、中国保险行业协会\r\n> **声明**: 以上动态供参考，具体以官方最新发布为准\r\n\r\n\r\n\r\n> **English:** AI-powered insurance agent coaching system — parses product documents, generates\r\n> personalized question banks, assesses agent competency levels, schedules daily training based on\r\n> client visits, and runs interactive role-play drills. Benchmarked against AIA, Ping An, and\r\n> Alibaba Cloud insurance training systems.\r\n>\r\n> **中文:** 保险代理人智能陪练系统——解析产品文档、自动生成问题库、评估代理人能力等级、\r\n> 结合当日客户拜访行程安排个性化训练、进行情景对练。对标友邦保险、平安保险、阿里云智能陪练水平。\r\n\r\n---\r\n\r\n## Trigger Keywords / 触发关键词\r\n\r\n**⚠️ 精确触发规则**：仅当用户明确提到保险代理人培训/陪练相关需求时激活。日常对话中提及\"培训\"、\"训练\"、\"coaching\"、\"agent training\"等通用词汇时**不会自动触发**。\r\n\r\n**用户确认规则**：当用户输入匹配以下关键词时，必须先确认用户意图：\r\n- \"您需要保险代理人陪练/培训服务吗？\"\r\n- 仅在用户明确确认后，才进入陪练模式\r\n\r\n激活关键词（需用户确认后生效）：\r\n\r\n- 保险陪练 / 产品陪练 / 智能陪练 / 代理人训练\r\n- 代理人培训 / 新人培训 / 保险话术训练\r\n- 产品演练 / 客户异议处理 / 保险销售训练\r\n- insurance agent training / insurance coaching / insurance product drill\r\n\r\n---\r\n\r\n## Core System Architecture / 核心系统架构\r\n\r\n### 0. 2025-2026 代理人销售环境最新变化\r\n\r\n| 变化 | 内容 | 话术调整建议 |\r\n|------|------|------------|\r\n| **预定利率降至3.0%** | 2024年9月后所有新产品执行 | 强调\"锁定3.0%长期确定收益\"，对比银行理财波动性 |\r\n| **分红险主导市场** | 分红险、万能险替代传统高利率产品 | 学会讲\"浮动收益+保底保障\"的双重价值 |\r\n| **健康险新规上线** | 2025年商业健康险管理办法修订 | 健康告知流程需更规范，禁止误导性说明 |\r\n| **代理人资格考试升级** | 2025年加入AI伦理、数字化服务模块 | 新人需补充数字化能力培训 |\r\n| **企微客户触达合规** | AI外呼需标注身份，营销需客户授权 | 培训合规营销话术，避免违规外呼 |\r\n\r\n\r\n\r\n```\r\n┌─────────────────────────────────────────────────────────────────┐\r\n│ Insurance Agent Intelligent Trainer │\r\n├─────────────────────────────────────────────────────────────────┤\r\n│ ┌──────────────┐ ┌──────────────┐ ┌──────────────────────┐ │\r\n│ │ Product Doc │ │ Agent Profile│ │ Daily Schedule/Routes│ │\r\n│ │ Parser │ │ Engine │ │ Integration │ │\r\n│ │ (PDF/Word/ │ │ (Skill Level │ │ (Today's Visits & │ │\r\n│ │ Images) │ │ Assessment) │ │ Client Profiles) │ │\r\n│ └──────┬───────┘ └──────┬───────┘ └──────────┬───────────┘ │\r\n│ │ │ │ │\r\n│ ▼ ▼ ▼ │\r\n│ ┌──────────────────────────────────────────────────────────┐ │\r\n│ │ Question Bank Generation Engine │ │\r\n│ │ Product Knowledge │ Objection Handling │ Case Analysis │ │\r\n│ │ [5 difficulty tiers × 3 categories = 15 question types] │ │\r\n│ └──────────────────────────┬───────────────────────────────┘ │\r\n│ │ │\r\n│ ▼ │\r\n│ ┌──────────────────────────────────────────────────────────┐ │\r\n│ │ Personalized Training Scheduler │ │\r\n│ │ [Skill Level + Schedule + Product Priority = Daily Plan]│ │\r\n│ └──────────────────────────┬───────────────────────────────┘ │\r\n│ │ │\r\n│ ▼ │\r\n│ ┌──────────────────────────────────────────────────────────┐ │\r\n│ │ Interactive Training Engine │ │\r\n│ │ Role-play │ Real-time Feedback │ Progress Tracking │ │\r\n│ └──────────────────────────────────────────────────────────┘ │\r\n└─────────────────────────────────────────────────────────────────┘\r\n```\r\n\r\n---\r\n\r\n\r\n### 保险监管最新动态 [2026-06-28更新]\r\n\r\n| 动态类型 | 内容摘要 | 发布时间 | 影响范围 |\r\n|---------|---------|---------|---------|\r\n| 监管发布 | 金融监管总局发布《关于银行业保险业人工智能安全开发应用的指导意见》，界定承保理赔、风险管理等为AI高风险应用场景 | 2026-06-18 | 保险AI应用合规与风险管控 |\r\n| 监管计划 | NFRA发布《2026年规章制定工作计划》：保险资金运用管理办法、偿付能力管理办法纳入修订，流动性风险、网络安全管理办法新制定 | 2026-06-23 | 保险监管全链条 |\r\n| 监管施行 | NFRA 2026年第2号令《银行保险机构许可证管理办法》6月1日起施行，取消保险许可证统一为金融许可证 | 2026-06-01 | 保险业务资质与合规管理 |\r\n\r\n> **数据截止**: 2026-06-28 | 来源：国家金融监督管理总局、行业公开信息\r\n> **声明**: 以上动态供参考，具体以官方最新发布为准\r\n\r\n## Core Capabilities / 核心能力\r\n\r\n### 1. Product Document Parser / 产品文档解析引擎（教学演示）\r\n\r\n> **⚠️ 教学演示**：以下展示产品文档解析的**概念性教学方法论**，仅说明AI可如何辅助理解产品结构。**本技能不执行任何实际的PDF解析、OCR识别或文档提取操作。** 所有\"解析流程\"均为逻辑示意，实际应用需由具体的工程实现完成。\r\n\r\n**Supported formats (conceptual):** PDF, Word (.docx), scanned images (with OCR), plain text\r\n\r\n**Conceptual parsing pipeline (for reference):**\r\n\r\n```\r\nDocument Upload\r\n │\r\n ▼\r\n[Format Detection] → PDF / Word / Image / Text\r\n │\r\n ▼\r\n[Text Extraction] → Raw text content\r\n │\r\n ▼\r\n[Structure Analysis]\r\n ├─ Product name, type, target customers\r\n ├─ Coverage scope (death, medical, annuity, critical illness, etc.)\r\n ├─ Premium levels & payment periods\r\n ├─ Policy terms & exclusions\r\n ├─ Sales pitch key points\r\n ├─ Competitive advantages vs. similar products\r\n └─ Compliance notes & regulatory requirements\r\n │\r\n ▼\r\n[Structured Product Profile] → Ready for question generation\r\n```\r\n\r\n**Output: Structured Product Profile JSON**\r\n\r\n```json\r\n{\r\n \"product_name\": \"XX福享人生终身寿险(万能型)\",\r\n \"product_type\": \"whole-life insurance with universal account\",\r\n \"insurer\": \"国联人寿\",\r\n \"target_customers\": [\"30-50岁中高收入人群\", \"有财富传承需求\"],\r\n \"coverage\": {\r\n \"death_benefit\": \"100%-160%账户价值\",\r\n \"annuity_option\": \"60岁起可转换为年金\",\r\n \"waiver\": \"可选投保人保费豁免\"\r\n },\r\n \"premium\": {\r\n \"min_annual\": 12000,\r\n \"payment_periods\": [\"3年\", \"5年\", \"10年\", \"20年\"],\r\n \"min_coverage_years\": \"终身\"\r\n },\r\n \"key_selling_points\": [\r\n \"复利增值，万能账户历史结算利率4.5%-5.2%\",\r\n \"灵活追加，额外资金可随时进入万能账户\",\r\n \"身故保障与财富传承双重功能\"\r\n ],\r\n \"competitive_edges\": [\"结算利率优于同类竞品\", \"追加无上限\"],\r\n \"exclusions\": [\"投保人对被保险人的故意伤害\", \"2年内自杀(无民事行为能力人除外)\"],\r\n \"compliance_notes\": [\"需双录(录音录像)\", \"犹豫期15天\", \"等待期90天\"],\r\n \"difficulty_tags\": [\"新人友好\", \"需强化健康告知\", \"财务规划综合能力\"]\r\n}\r\n```\r\n\r\n---\r\n\r\n### 2. Agent Profile & Skill Assessment / 代理人画像与能力评估\r\n\r\n> **⚠️ 数据处理提醒**：以下代理人画像和日程数据为**演示示例**。实际使用时，用户应自行管理代理人数据的收集和存储，确保符合《个人信息保护法》及保险行业合规要求。请勿输入真实客户PII信息。\r\n\r\n**Three skill tiers:**\r\n\r\n| Tier | Level | Description | Training Focus |\r\n|------|-------|-------------|----------------|\r\n| 🌱 **L1 - 入门级** | Beginner | < 1 year experience, struggles with product details and objection handling | Foundation: product knowledge, basic sales scripts, simple objection responses |\r\n| ⚡ **L2 - 进阶级** | Intermediate | 1-3 years, solid product knowledge but inconsistent closing rate | Application: complex scenarios, multi-product combination, competitive replacement, high-net-worth clients |\r\n| 🎯 **L3 - 专家级** | Advanced | 3+ years, high performance, needs strategy for complex cases | Mastery: enterprise/group clients, tax planning, estate planning, competitive stealing, mentoring skills |\r\n\r\n**Profile structure:**\r\n\r\n```json\r\n{\r\n \"agent_id\": \"AG20240001\",\r\n \"name\": \"张明\",\r\n \"level\": \"L2\",\r\n \"level_label\": \"进阶级\",\r\n \"tenure_years\": 2.5,\r\n \"certifications\": [\"保险代理人资格证\", \"健康险销售资质\"],\r\n \"performance\": {\r\n \"monthly_premium_target\": 50000,\r\n \"monthly_premium_actual\": 42000,\r\n \"closing_rate\": 0.32,\r\n \"avg_policy_size\": 18500,\r\n \"new_customer_rate\": 0.45\r\n },\r\n \"product_mastery\": {\r\n \"term_life\": 0.85,\r\n \"whole_life\": 0.72,\r\n \"critical_illness\": 0.58,\r\n \"medical_insurance\": 0.80,\r\n \"annuity\": 0.45,\r\n \"investment_linked\": 0.38\r\n },\r\n \"weak_points\": [\r\n \"健康险异议处理不够熟练\",\r\n \"不了解高端客户的税务筹划需求\",\r\n \"组合产品销售话术单一\"\r\n ],\r\n \"strong_points\": [\r\n \"老客户维护能力强\",\r\n \"缘故市场开拓优秀\"\r\n ],\r\n \"daily_schedule\": [\r\n {\"time\": \"09:00-10:00\", \"activity\": \"晨会\", \"location\": \"营业部\"},\r\n {\"time\": \"10:30-12:00\", \"activity\": \"拜访客户A（国企中层，有养老需求）\", \"location\": \"客户公司\"},\r\n {\"time\": \"14:00-15:30\", \"activity\": \"拜访客户B（私企业主，健康险需求）\", \"location\": \"客户公司\"},\r\n {\"time\": \"16:00-17:30\", \"activity\": \"缘故客户C（教育金规划）\", \"location\": \"咖啡厅\"}\r\n ]\r\n}\r\n```\r\n\r\n---\r\n\r\n### 3. Question Bank Generation / 问题库自动生成（教学模板）\r\n\r\n> **⚠️ 教学演示**：以下问题库和话术为**培训场景的教学参考模板**，展示如何结构化设计代理人训练内容。所有涉及销售话术、竞品对比、异议处理的内容均为**培训素材**，实际销售行为须遵循《保险法》及相关监管规定，并经持牌保险专业人士审核后方可执行。\r\n\r\n**Generated from product profile + agent level + training objectives**\r\n\r\n#### Question Types (15 categories across 3 dimensions)\r\n\r\n**By Category:**\r\n\r\n| Category | Description | Example |\r\n|----------|-------------|---------|\r\n| **产品知识** | Product features, terms, coverage | \"XX福的等待期是多久？\" |\r\n| **客户画像** | Target customer identification | \"什么样的客户适合购买这款产品？\" |\r\n| **异议处理** | Objection handling scripts | \"客户说'我已经有社保了，不需要商业保险'，如何回应？\" |\r\n| **案例分析** | Real case discussion | \"40岁国企中层，年薪50万，如何用这款产品做养老规划？\" |\r\n| **合规话术** | Compliance-approved scripts | \"如何向客户解释犹豫期和退保损失？\" |\r\n| **竞品对比** | vs. competitors | \"相比平安福，这款产品的核心优势是什么？\" |\r\n| **促成话术** | Closing techniques | \"客户表现出购买意向，如何自然促成？\" |\r\n| **交叉销售** | Multi-product combination | \"如何将主险与医疗险组合销售？\" |\r\n\r\n**By Difficulty (5 tiers):**\r\n\r\n| Level | Target Audience | Question Complexity |\r\n|--------|----------------|---------------------|\r\n| ⭐ 基础 | L1新人 | 单一产品，单一问题，直接答案 |\r\n| ⭐⭐ 入门 | L1-L2 | 单一产品，1-2个知识点，需要解释 |\r\n| ⭐⭐⭐ 进阶 | L2 | 单一产品，3-5个知识点，需组合分析 |\r\n| ⭐⭐⭐⭐ 高阶 | L2-L3 | 多产品组合，竞争替换，高净值客户 |\r\n| ⭐⭐⭐⭐⭐ 专家 | L3 | 综合方案，税务筹划，财富传承 |\r\n\r\n#### Question Bank Generation Prompt:\r\n\r\n```\r\nBased on the product profile provided, generate a question bank with:\r\n\r\n1. For each difficulty tier (基础/入门/进阶/高阶/专家):\r\n - 5 multiple choice questions (产品知识)\r\n - 3 case analysis questions\r\n - 3 objection handling scenarios\r\n - 2 competitive comparison questions\r\n - 1 closing technique exercise\r\n\r\n2. Total: 65+ questions per product\r\n\r\n3. For each question, provide:\r\n - Question text\r\n - Difficulty level (1-5)\r\n - Category (产品知识/异议处理/案例分析/竞品对比/促成话术)\r\n - Ideal answer / model response\r\n - Evaluation criteria (excellent/good/needs-improvement)\r\n - Coaching tips for the trainer\r\n```\r\n\r\n---\r\n\r\n### 4. Personalized Training Scheduler / 个性化训练调度引擎（方法论演示）\r\n\r\n> **⚠️ 教学演示**：以下调度算法、代理人画像及日程数据均为**教学方法论的概念性展示**。**本技能不实际采集、存储或处理任何代理人或客户数据**。所有姓名、日程、业绩数据均为虚构示例，仅用于说明逻辑框架。\r\n\r\n**Input factors:**\r\n\r\n```\r\nAgent Profile (Level + Weak Points)\r\n +\r\nToday's Client Schedule (Who → What need → What product)\r\n +\r\nProduct Priority Matrix\r\n =\r\nPersonalized Daily Training Plan\r\n```\r\n\r\n**Scheduling Algorithm:**\r\n\r\n```python\r\ndef generate_daily_training_plan(agent_profile, daily_schedule, products):\r\n \"\"\"\r\n Generate personalized training plan for the day.\r\n \"\"\"\r\n # Step 1: Identify today's client visit products\r\n today_products = extract_products_from_schedule(daily_schedule)\r\n \r\n # Step 2: Get agent's weakness areas for these products\r\n weakness_map = get_weakness_for_products(\r\n agent_profile.weak_points, \r\n today_products\r\n )\r\n \r\n # Step 3: Calculate training time available\r\n available_minutes = calculate_available_training_time(daily_schedule)\r\n \r\n # Step 4: Prioritize by impact × weakness × product value\r\n training_queue = prioritize_training(\r\n weakness_map,\r\n today_products,\r\n agent_profile.level,\r\n time_constraint=available_minutes\r\n )\r\n \r\n # Step 5: Generate session plan\r\n sessions = split_into_sessions(training_queue, available_minutes)\r\n \r\n return {\r\n \"date\": today,\r\n \"agent\": agent_profile.name,\r\n \"total_minutes\": available_minutes,\r\n \"sessions\": sessions,\r\n \"focus_products\": today_products,\r\n \"key_objectives\": get_key_objectives(training_queue)\r\n }\r\n```\r\n\r\n**Example Daily Training Plan:**\r\n\r\n```json\r\n{\r\n \"date\": \"2026-05-05\",\r\n \"agent\": \"张明\",\r\n \"level\": \"L2\",\r\n \"total_minutes\": 90,\r\n \"sessions\": [\r\n {\r\n \"time\": \"08:00-08:20\",\r\n \"duration\": 20,\r\n \"type\": \"晨间快练\",\r\n \"mode\": \"快问快答\",\r\n \"focus\": \"年金险产品知识（高频问题5题）\",\r\n \"product\": \"福享人生终身寿险\",\r\n \"objective\": \"巩固年金转换权的计算逻辑\"\r\n },\r\n {\r\n \"time\": \"12:30-13:00\",\r\n \"duration\": 30,\r\n \"type\": \"午间强化\",\r\n \"mode\": \"情景对练\",\r\n \"focus\": \"健康险异议处理\",\r\n \"scenario\": \"客户：\"我有社保，不需要商业医疗险\"\",\r\n \"product\": \"康健医疗保险\",\r\n \"level\": \"⭐⭐⭐ 进阶\",\r\n \"coaching_tips\": \"引导客户认识到社保报销比例上限，用自费药比例对比引发需求\"\r\n },\r\n {\r\n \"time\": \"17:30-18:30\",\r\n \"duration\": 40,\r\n \"type\": \"晚间复盘\",\r\n \"mode\": \"案例分析 + 角色扮演\",\r\n \"focus\": \"私企业主综合保障方案\",\r\n \"scenario\": \"45岁私企老板，年收入200万，已有多份保单，如何做加保方案？\",\r\n \"products\": [\"终身寿险+万能账户\", \"高端医疗\", \"企业财产险\"],\r\n \"level\": \"⭐⭐⭐⭐ 高阶\",\r\n \"model_response_guide\": \"从家庭资产与企业资产隔离角度切入，引出终身寿险的债务隔离和传承功能\"\r\n }\r\n ],\r\n \"key_metrics_to_track\": [\r\n \"异议处理响应时间（目标<30秒）\",\r\n \"产品知识点正确率（目标>85%）\",\r\n \"方案组合完整性（3单以上产品覆盖）\"\r\n ]\r\n}\r\n```\r\n\r\n---\r\n\r\n### 5. Interactive Training Session / 智能陪练对话引擎\r\n\r\n**Session modes:**\r\n\r\n| Mode | Description | Duration | Best For |\r\n|------|-------------|----------|----------|\r\n| **快问快答** | Rapid-fire Q&A | 5-10 min | Pre-meeting warmup |\r\n| **情景对练** | Role-play (client vs. agent) | 15-30 min | Skill practice |\r\n| **案例研讨** | Real case analysis | 20-40 min | Advanced agents |\r\n| **异议攻关** | Objection busting focus | 10-15 min | Weak point training |\r\n| **综合考核** | Full simulation exam | 30-60 min | Level assessment |\r\n\r\n**Real-time coaching during training:**\r\n\r\n```\r\nAgent Response\r\n │\r\n ▼\r\n[Natural Language Understanding] → Extract key claims, tone, strategy\r\n │\r\n ▼\r\n[Evaluation Engine]\r\n ├─ Product knowledge accuracy ✓/✗\r\n ├─ Objection handling effectiveness (1-5)\r\n ├─ Compliance adherence ✓/✗\r\n ├─ Closing attempt timing (good/early/late/missing)\r\n ├─ Client empathy signals ✓/✗\r\n └─ Product combination logic ✓/✗\r\n │\r\n ▼\r\n[Real-time Coaching Feedback]\r\n ├─ Immediate tip (if struggling): \"💡 提示：可以先问客户目前的保障缺口...\"\r\n ├─ Completion praise (if excellent): \"🌟 完美！您已经很好地识别了客户需求\"\r\n └─ Post-question summary: \"本轮得分 85/100。建议加强：竞品对比环节\"\r\n```\r\n\r\n**Training session flow:**\r\n\r\n```\r\n1. 导入 (5%) → 介绍训练目标和产品背景\r\n2. 暖场 (10%) → 快问快答热身，激活产品知识\r\n3. 主体 (60%) → 情景对练：客户角色扮演 + 实时点评\r\n4. 复盘 (20%) → AI给出详细反馈：优点/不足/改进建议\r\n5. 行动 (5%) → 下次拜访的具体行动计划\r\n```\r\n\r\n---\r\n\r\n### 6. Effect Assessment & Progress Tracking / 效果评估与进度追踪\r\n\r\n**Metrics tracked per session:**\r\n\r\n| Metric | Definition | Target |\r\n|--------|------------|--------|\r\n| **产品知识得分** | 知识点正确率 | L1: ≥70%, L2: ≥80%, L3: ≥90% |\r\n| **异议处理时效** | 从异议提出到满意回答的时间 | < 30秒 |\r\n| **促成成功率** | 能否自然引入促成信号 | ≥ 1次有效尝试 |\r\n| **话术合规率** | 合规敏感词使用正确性 | 100% |\r\n| **方案完整性** | 保障覆盖广度 | ≥ 3个维度 |\r\n\r\n**Progress report structure:**\r\n\r\n```markdown\r\n## 📊 代理人张明 训练报告 - 2026-05-05\r\n\r\n### 综合得分: ⭐⭐⭐⭐ (78/100)\r\n\r\n| 维度 | 本次得分 | 较上次 | 目标 |\r\n|------|---------|--------|------|\r\n| 产品知识 | 82/100 | ↑5 | 80+ |\r\n| 异议处理 | 71/100 | ↓3 | 75+ |\r\n| 促成技巧 | 85/100 | ↑8 | 80+ |\r\n| 合规话术 | 95/100 | →0 | 100 |\r\n| 方案设计 | 72/100 | ↑12 | 75+ |\r\n\r\n### 🔥 本次表现亮点\r\n1. 养老规划方案逻辑清晰，能结合客户生命周期讲解\r\n2. 合规话术使用规范，犹豫期/退保说明完整\r\n\r\n### ⚠️ 需要加强\r\n1. 健康险异议处理：回应\"已有社保\"时过于被动，应主动算账\r\n2. 竞品对比：对中国平安主要产品线不够熟悉\r\n\r\n### 📅 明日训练重点\r\n- 产品：康健医疗保险（健康告知流程）\r\n- 场景：竞品替换（平安福 vs. XX福）\r\n- 时长：30分钟情景对练 + 10分钟快问快答\r\n```\r\n\r\n---\r\n\r\n## Workflow / 标准工作流程\r\n\r\n> **⚠️ 重要提示**：以下工作流展示的是**培训场景的教学参考**。所有销售话术和异议处理内容均为培训素材，实际销售行为须遵循《保险法》及相关监管规定，经持牌保险专业人士审核。\r\n\r\n### Mode 1: Quick Start (已知产品 + 快速训练)\r\n\r\n```\r\nUser: \"帮我准备明天拜访客户B的训练，他是私企老板，对健康险感兴趣\"\r\n │\r\n ▼\r\n[Step 1] 获取代理人信息 → 张明，L2，弱项：健康险异议处理\r\n[Step 2] 识别拜访产品 → 康健医疗保险（目标：替换平安福）\r\n[Step 3] 生成训练计划 → 午间30分钟：健康险异议处理对练\r\n[Step 4] 开始陪练 → 情景对练：私企业主健康险需求挖掘\r\n[Step 5] 实时反馈 → 异议处理评分：71/100，给出改进建议\r\n[Step 6] 报告输出 → 训练报告 + 明日拜访话术优化建议\r\n```\r\n\r\n### Mode 2: Product Document Upload (上传产品文档)\r\n\r\n```\r\nUser: [上传 XX保险公司福享人生终身寿险 产品手册 PDF]\r\n │\r\n ▼\r\n[Step 1] 解析文档 → 提取产品结构、条款、卖点\r\n[Step 2] 生成产品画像 → Structured JSON Profile\r\n[Step 3] 生成问题库 → 65+道题目（5难度×8类别）\r\n[Step 4] 生成参考题库 → 作为AI对话上下文，**不持久存储**\r\n[Step 5] 等待选择 → \"请选择训练模式：快问快答 / 情景对练 / 案例研讨\"\r\n```\r\n\r\n### Mode 3: Full Agent Assessment (全面能力评估)\r\n\r\n```\r\nUser: \"帮我评估代理人李华的综合能力，她入职8个月，主要卖重疾险\"\r\n │\r\n ▼\r\n[Step 1] 建立代理人档案 → L1入门级，8个月，重疾险方向\r\n[Step 2] 产品文档上传 → 重疾险产品手册\r\n[Step 3] 综合考核 → 30题产品知识 + 5个情景对练\r\n[Step 4] 生成能力雷达图 → 6维度能力可视化\r\n[Step 5] 制定成长路径 → 90天训练计划\r\n```\r\n\r\n---\r\n\r\n## Input / Output Specifications / 输入输出规范\r\n\r\n### Input\r\n\r\n| Input Type | Description | Example |\r\n|------------|-------------|---------|\r\n| 代理人档案 | JSON/文本描述 | 姓名、级别、工龄、业绩、弱项 |\r\n| 产品文档 | PDF/Word/TXT/图片 | 保险产品手册、条款、计划书 |\r\n| 当日行程 | 文本/日历 | 09:00晨会 / 10:30拜访客户A |\r\n| 训练指令 | 自然语言 | \"帮我准备健康险的陪练\" |\r\n| 客户信息 | 文本描述 | \"45岁私企老板，年收入200万\" |\r\n\r\n### Output\r\n\r\n| Output Type | Description |\r\n|-------------|-------------|\r\n| 产品画像JSON | 结构化产品信息 |\r\n| 问题库 | 65+道分类分级题目 |\r\n| 训练计划 | 分钟级个性化日程 |\r\n| 陪练对话 | 实时AI角色扮演 |\r\n| 评估报告 | 评分 + 改进建议 + 雷达图 |\r\n| 成长路径 | 30/60/90天训练建议 |\r\n\r\n---\r\n\r\n## Integration Notes / 集成说明\r\n\r\n**Data privacy:**\r\n- All agent and client data remains local / within the company's system\r\n- No sensitive PII should be included in training documents\r\n- Comply with China CBIRC insurance sales compliance regulations\r\n\r\n**Lianxi with other Skills:**\r\n- `insurance-bidding-pro`: Use product analysis for bidding scenarios\r\n- `insurance-private-domain-ops`: Link training completion to customer follow-up\r\n- `insurance-claims-intelligence`: Train agents on claim processes for better client communication\r\n\r\n---\r\n\r\n## Disclaimer / 免责声明\r\n\r\n> ⚠️ **Training is advisory only.** This skill provides coaching materials, question banks,\r\n> and simulation training for insurance agent development. All final sales advice,\r\n> compliance decisions, and product recommendations must be reviewed by licensed\r\n> insurance professionals and comply with CBIRC regulations. Model answers represent\r\n> reference best practices, not guaranteed outcomes.\n\nFile v5.2.3:_meta.json\n\n{\n \"ownerId\": \"kn74e704j3ygjcygnpf02rdvd185js13\",\n \"slug\": \"insurance-agent-trainer\",\n \"version\": \"5.2.3\",\n \"publishedAt\": 1784386900423\n}\n\nFile v5.2.3:skill-card.md\n\n## Description:
\nAI-powered insurance agent training coach that generates question banks, skill assessments, personalized training plans, and role-play practice for insurance-agent development.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[gechengling](https://clawhub.ai/user/gechengling)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nInsurance training teams and agents use this skill to create training content, assessment prompts, role-play scenarios, and daily practice plans for insurance sales development. Generated guidance is for training support and should be reviewed by licensed insurance or compliance staff before real-world use.
\n\n### Deployment Geography for Use:
\nGlobal, with China-specific regulatory examples and compliance references.
\n\n## Known Risks and Mitigations:
\nRisk: Generated insurance scripts, product comparisons, and regulatory claims may be incomplete or inaccurate if used as compliance or sales advice.
\nMitigation: Have licensed insurance or compliance staff review generated materials before use with agents or customers.
\nRisk: Users may provide real customer PII or sensitive business data while preparing training scenarios.
\nMitigation: Use approved local handling controls and avoid entering real customer PII unless the organization has authorized that workflow.
\nRisk: Document-upload parsing claims may be misunderstood as a verified document-processing implementation.
\nMitigation: Treat document parsing behavior as a training-content framework capability unless separately validated in the deployment environment.
\n\n\n## Reference(s):
\n- [ClawHub skill page](https://clawhub.ai/gechengling/skills/insurance-agent-trainer)
\n\n\n## Skill Output:
\n**Output Type(s):** [text, markdown, code, configuration, guidance]
\n**Output Format:** [Markdown and structured text with example JSON and Python snippets]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [Produces training materials, question banks, role-play prompts, assessment reports, and practice-plan guidance that require human review.]
\n\n## Skill Version(s):
\n5.2.3 (source: server evidence and frontmatter)
\n\n## Ethical Considerations:
\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.
\n\nArchive v5.2.2: 10 files, 41594 bytes\n\nFiles: README.md (6858b), references/agent_profile_template.md (9635b), references/question_bank_templates.md (13893b), references/training_evaluation_rubric.md (6398b), scripts/product_parser.py (8626b), scripts/question_generator.py (9212b), scripts/training_scheduler.py (11972b), skill-card.md (3023b), SKILL.md (28116b), _meta.json (142b)\n\nFile v5.2.2:SKILL.md\n\n---\r\nname: Insurance Agent Intelligent Trainer\r\ndescription: >\r\n AI-powered insurance agent training coach — auto-parses product docs, generates question banks,\r\n assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training\r\n based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers\r\n predinig rate cut (3.0%) impact on sales scripts, new health insurance regulation, PIPL-compliant\r\n customer communication, and benchmark against AIA, Ping An, and Alibaba Cloud training systems.\r\n Keywords: 智能陪练, 代理人培训, 保险训练, 产品陪练, 话术训练, 角色扮演, 技能评估, AIA培训, 平安培训, 代理人展业, 保险销售, 客户面谈, 异议处理, 保险培训系统, 学习路径.\r\nslug: insurance-agent-trainer\r\nversion: 5.2.2\r\nallowed-tools: []\r\ncapabilities:\r\n - educational-reference\r\n - advisory-only\r\n - requires-human-review\r\n - no-executable-code\r\n---\r\n\r\n# Insurance Agent Intelligent Trainer / 保险代理人智能陪练系统\r\n\r\n> **⚠️ SECURITY NOTICE / 安全声明**\r\n> - **Type:** Educational reference / analytical framework ONLY\r\n> - **No executable code, scripts, or binaries are included in this skill**\r\n> - **No persistent storage, network calls, background execution, or credential collection**\r\n> - **All outputs are for reference only and require human review before real-world application**\r\n> - **This skill does NOT provide financial, legal, or insurance advice**\r\n> - **Users must exercise their own judgment and consult qualified professionals**\r\n>\r\n> **⚠️ 数据安全警告**\r\n> - 本技能仅提供保险代理人的培训辅导参考框架，**不执行任何代码或脚本**\r\n> - 所有文档解析、日程分析、画像评估的描述均为**教学参考框架**，**不包含实际的OCR或PDF解析引擎**\r\n> - 不会自动访问、存储或处理用户的任何培训数据或个人信息\r\n> - 培训计划和话术建议需结合用户实际业务场景调整，**不能替代专业培训师**\r\n> - **销售话术和异议处理仅为培训参考，实际使用须遵守《保险法》及相关监管规定，不得以AI输出替代合规审核**\r\n\r\n### 保险监管最新动态 [2026-06-15更新]\r\n\r\n| 动态类型 | 内容摘要 | 发布时间 | 影响范围 |\r\n|---------|---------|---------|---------|\r\n| 监管发布 | NFRA 2026年第2号令：《银行保险机构许可证管理办法》6月1日施行 | 2026-06 | 代理人培训需新增许可证管理知识模块 |\r\n| 监管发布 | 许可证换证过渡期2026.6-2028.5，换证流程纳入培训 | 2026-06 | 代理人换证操作培训 |\r\n| 监管动态 | 2026年Q1监管处罚个人追责条款落地 | 2026-Q1 | 代理人合规培训需强化个人责任意识 |\r\n\r\n> **数据截止**: 2026-06-15 | 来源：国家金融监督管理总局、政府网、金融新闻网\r\n> **声明**: 以上动态供参考，具体以官方最新发布为准\r\n\r\n\r\n\r\n> **English:** AI-powered insurance agent coaching system — parses product documents, generates\r\n> personalized question banks, assesses agent competency levels, schedules daily training based on\r\n> client visits, and runs interactive role-play drills. Benchmarked against AIA, Ping An, and\r\n> Alibaba Cloud insurance training systems.\r\n>\r\n> **中文:** 保险代理人智能陪练系统——解析产品文档、自动生成问题库、评估代理人能力等级、\r\n> 结合当日客户拜访行程安排个性化训练、进行情景对练。对标友邦保险、平安保险、阿里云智能陪练水平。\r\n\r\n---\r\n\r\n## Trigger Keywords / 触发关键词\r\n\r\n**⚠️ 精确触发规则**：仅当用户明确提到保险代理人培训/陪练相关需求时激活。日常对话中提及\"培训\"、\"训练\"、\"coaching\"、\"agent training\"等通用词汇时**不会自动触发**。\r\n\r\n**用户确认规则**：当用户输入匹配以下关键词时，必须先确认用户意图：\r\n- \"您需要保险代理人陪练/培训服务吗？\"\r\n- 仅在用户明确确认后，才进入陪练模式\r\n\r\n激活关键词（需用户确认后生效）：\r\n\r\n- 保险陪练 / 产品陪练 / 智能陪练 / 代理人训练\r\n- 代理人培训 / 新人培训 / 保险话术训练\r\n- 产品演练 / 客户异议处理 / 保险销售训练\r\n- insurance agent training / insurance coaching / insurance product drill\r\n\r\n---\r\n\r\n## Core System Architecture / 核心系统架构\r\n\r\n### 0. 2025-2026 代理人销售环境最新变化\r\n\r\n| 变化 | 内容 | 话术调整建议 |\r\n|------|------|------------|\r\n| **预定利率降至3.0%** | 2024年9月后所有新产品执行 | 强调\"锁定3.0%长期确定收益\"，对比银行理财波动性 |\r\n| **分红险主导市场** | 分红险、万能险替代传统高利率产品 | 学会讲\"浮动收益+保底保障\"的双重价值 |\r\n| **健康险新规上线** | 2025年商业健康险管理办法修订 | 健康告知流程需更规范，禁止误导性说明 |\r\n| **代理人资格考试升级** | 2025年加入AI伦理、数字化服务模块 | 新人需补充数字化能力培训 |\r\n| **企微客户触达合规** | AI外呼需标注身份，营销需客户授权 | 培训合规营销话术，避免违规外呼 |\r\n\r\n\r\n\r\n```\r\n┌─────────────────────────────────────────────────────────────────┐\r\n│ Insurance Agent Intelligent Trainer │\r\n├─────────────────────────────────────────────────────────────────┤\r\n│ ┌──────────────┐ ┌──────────────┐ ┌──────────────────────┐ │\r\n│ │ Product Doc │ │ Agent Profile│ │ Daily Schedule/Routes│ │\r\n│ │ Parser │ │ Engine │ │ Integration │ │\r\n│ │ (PDF/Word/ │ │ (Skill Level │ │ (Today's Visits & │ │\r\n│ │ Images) │ │ Assessment) │ │ Client Profiles) │ │\r\n│ └──────┬───────┘ └──────┬───────┘ └──────────┬───────────┘ │\r\n│ │ │ │ │\r\n│ ▼ ▼ ▼ │\r\n│ ┌──────────────────────────────────────────────────────────┐ │\r\n│ │ Question Bank Generation Engine │ │\r\n│ │ Product Knowledge │ Objection Handling │ Case Analysis │ │\r\n│ │ [5 difficulty tiers × 3 categories = 15 question types] │ │\r\n│ └──────────────────────────┬───────────────────────────────┘ │\r\n│ │ │\r\n│ ▼ │\r\n│ ┌──────────────────────────────────────────────────────────┐ │\r\n│ │ Personalized Training Scheduler │ │\r\n│ │ [Skill Level + Schedule + Product Priority = Daily Plan]│ │\r\n│ └──────────────────────────┬───────────────────────────────┘ │\r\n│ │ │\r\n│ ▼ │\r\n│ ┌──────────────────────────────────────────────────────────┐ │\r\n│ │ Interactive Training Engine │ │\r\n│ │ Role-play │ Real-time Feedback │ Progress Tracking │ │\r\n│ └──────────────────────────────────────────────────────────┘ │\r\n└─────────────────────────────────────────────────────────────────┘\r\n```\r\n\r\n---\r\n\r\n\r\n### 保险监管最新动态 [2026-06-28更新]\r\n\r\n| 动态类型 | 内容摘要 | 发布时间 | 影响范围 |\r\n|---------|---------|---------|---------|\r\n| 监管发布 | 金融监管总局发布《关于银行业保险业人工智能安全开发应用的指导意见》，界定承保理赔、风险管理等为AI高风险应用场景 | 2026-06-18 | 保险AI应用合规与风险管控 |\r\n| 监管计划 | NFRA发布《2026年规章制定工作计划》：保险资金运用管理办法、偿付能力管理办法纳入修订，流动性风险、网络安全管理办法新制定 | 2026-06-23 | 保险监管全链条 |\r\n| 监管施行 | NFRA 2026年第2号令《银行保险机构许可证管理办法》6月1日起施行，取消保险许可证统一为金融许可证 | 2026-06-01 | 保险业务资质与合规管理 |\r\n\r\n> **数据截止**: 2026-06-28 | 来源：国家金融监督管理总局、行业公开信息\r\n> **声明**: 以上动态供参考，具体以官方最新发布为准\r\n\r\n## Core Capabilities / 核心能力\r\n\r\n### 1. Product Document Parser / 产品文档解析引擎（教学演示）\r\n\r\n> **⚠️ 教学演示**：以下展示产品文档解析的**概念性教学方法论**，仅说明AI可如何辅助理解产品结构。**本技能不执行任何实际的PDF解析、OCR识别或文档提取操作。** 所有\"解析流程\"均为逻辑示意，实际应用需由具体的工程实现完成。\r\n\r\n**Supported formats (conceptual):** PDF, Word (.docx), scanned images (with OCR), plain text\r\n\r\n**Conceptual parsing pipeline (for reference):**\r\n\r\n```\r\nDocument Upload\r\n │\r\n ▼\r\n[Format Detection] → PDF / Word / Image / Text\r\n │\r\n ▼\r\n[Text Extraction] → Raw text content\r\n │\r\n ▼\r\n[Structure Analysis]\r\n ├─ Product name, type, target customers\r\n ├─ Coverage scope (death, medical, annuity, critical illness, etc.)\r\n ├─ Premium levels & payment periods\r\n ├─ Policy terms & exclusions\r\n ├─ Sales pitch key points\r\n ├─ Competitive advantages vs. similar products\r\n └─ Compliance notes & regulatory requirements\r\n │\r\n ▼\r\n[Structured Product Profile] → Ready for question generation\r\n```\r\n\r\n**Output: Structured Product Profile JSON**\r\n\r\n```json\r\n{\r\n \"product_name\": \"XX福享人生终身寿险(万能型)\",\r\n \"product_type\": \"whole-life insurance with universal account\",\r\n \"insurer\": \"国联人寿\",\r\n \"target_customers\": [\"30-50岁中高收入人群\", \"有财富传承需求\"],\r\n \"coverage\": {\r\n \"death_benefit\": \"100%-160%账户价值\",\r\n \"annuity_option\": \"60岁起可转换为年金\",\r\n \"waiver\": \"可选投保人保费豁免\"\r\n },\r\n \"premium\": {\r\n \"min_annual\": 12000,\r\n \"payment_periods\": [\"3年\", \"5年\", \"10年\", \"20年\"],\r\n \"min_coverage_years\": \"终身\"\r\n },\r\n \"key_selling_points\": [\r\n \"复利增值，万能账户历史结算利率4.5%-5.2%\",\r\n \"灵活追加，额外资金可随时进入万能账户\",\r\n \"身故保障与财富传承双重功能\"\r\n ],\r\n \"competitive_edges\": [\"结算利率优于同类竞品\", \"追加无上限\"],\r\n \"exclusions\": [\"投保人对被保险人的故意伤害\", \"2年内自杀(无民事行为能力人除外)\"],\r\n \"compliance_notes\": [\"需双录(录音录像)\", \"犹豫期15天\", \"等待期90天\"],\r\n \"difficulty_tags\": [\"新人友好\", \"需强化健康告知\", \"财务规划综合能力\"]\r\n}\r\n```\r\n\r\n---\r\n\r\n### 2. Agent Profile & Skill Assessment / 代理人画像与能力评估\r\n\r\n> **⚠️ 数据处理提醒**：以下代理人画像和日程数据为**演示示例**。实际使用时，用户应自行管理代理人数据的收集和存储，确保符合《个人信息保护法》及保险行业合规要求。请勿输入真实客户PII信息。\r\n\r\n**Three skill tiers:**\r\n\r\n| Tier | Level | Description | Training Focus |\r\n|------|-------|-------------|----------------|\r\n| 🌱 **L1 - 入门级** | Beginner | < 1 year experience, struggles with product details and objection handling | Foundation: product knowledge, basic sales scripts, simple objection responses |\r\n| ⚡ **L2 - 进阶级** | Intermediate | 1-3 years, solid product knowledge but inconsistent closing rate | Application: complex scenarios, multi-product combination, competitive replacement, high-net-worth clients |\r\n| 🎯 **L3 - 专家级** | Advanced | 3+ years, high performance, needs strategy for complex cases | Mastery: enterprise/group clients, tax planning, estate planning, competitive stealing, mentoring skills |\r\n\r\n**Profile structure:**\r\n\r\n```json\r\n{\r\n \"agent_id\": \"AG20240001\",\r\n \"name\": \"张明\",\r\n \"level\": \"L2\",\r\n \"level_label\": \"进阶级\",\r\n \"tenure_years\": 2.5,\r\n \"certifications\": [\"保险代理人资格证\", \"健康险销售资质\"],\r\n \"performance\": {\r\n \"monthly_premium_target\": 50000,\r\n \"monthly_premium_actual\": 42000,\r\n \"closing_rate\": 0.32,\r\n \"avg_policy_size\": 18500,\r\n \"new_customer_rate\": 0.45\r\n },\r\n \"product_mastery\": {\r\n \"term_life\": 0.85,\r\n \"whole_life\": 0.72,\r\n \"critical_illness\": 0.58,\r\n \"medical_insurance\": 0.80,\r\n \"annuity\": 0.45,\r\n \"investment_linked\": 0.38\r\n },\r\n \"weak_points\": [\r\n \"健康险异议处理不够熟练\",\r\n \"不了解高端客户的税务筹划需求\",\r\n \"组合产品销售话术单一\"\r\n ],\r\n \"strong_points\": [\r\n \"老客户维护能力强\",\r\n \"缘故市场开拓优秀\"\r\n ],\r\n \"daily_schedule\": [\r\n {\"time\": \"09:00-10:00\", \"activity\": \"晨会\", \"location\": \"营业部\"},\r\n {\"time\": \"10:30-12:00\", \"activity\": \"拜访客户A（国企中层，有养老需求）\", \"location\": \"客户公司\"},\r\n {\"time\": \"14:00-15:30\", \"activity\": \"拜访客户B（私企业主，健康险需求）\", \"location\": \"客户公司\"},\r\n {\"time\": \"16:00-17:30\", \"activity\": \"缘故客户C（教育金规划）\", \"location\": \"咖啡厅\"}\r\n ]\r\n}\r\n```\r\n\r\n---\r\n\r\n### 3. Question Bank Generation / 问题库自动生成（教学模板）\r\n\r\n> **⚠️ 教学演示**：以下问题库和话术为**培训场景的教学参考模板**，展示如何结构化设计代理人训练内容。所有涉及销售话术、竞品对比、异议处理的内容均为**培训素材**，实际销售行为须遵循《保险法》及相关监管规定，并经持牌保险专业人士审核后方可执行。\r\n\r\n**Generated from product profile + agent level + training objectives**\r\n\r\n#### Question Types (15 categories across 3 dimensions)\r\n\r\n**By Category:**\r\n\r\n| Category | Description | Example |\r\n|----------|-------------|---------|\r\n| **产品知识** | Product features, terms, coverage | \"XX福的等待期是多久？\" |\r\n| **客户画像** | Target customer identification | \"什么样的客户适合购买这款产品？\" |\r\n| **异议处理** | Objection handling scripts | \"客户说'我已经有社保了，不需要商业保险'，如何回应？\" |\r\n| **案例分析** | Real case discussion | \"40岁国企中层，年薪50万，如何用这款产品做养老规划？\" |\r\n| **合规话术** | Compliance-approved scripts | \"如何向客户解释犹豫期和退保损失？\" |\r\n| **竞品对比** | vs. competitors | \"相比平安福，这款产品的核心优势是什么？\" |\r\n| **促成话术** | Closing techniques | \"客户表现出购买意向，如何自然促成？\" |\r\n| **交叉销售** | Multi-product combination | \"如何将主险与医疗险组合销售？\" |\r\n\r\n**By Difficulty (5 tiers):**\r\n\r\n| Level | Target Audience | Question Complexity |\r\n|--------|----------------|---------------------|\r\n| ⭐ 基础 | L1新人 | 单一产品，单一问题，直接答案 |\r\n| ⭐⭐ 入门 | L1-L2 | 单一产品，1-2个知识点，需要解释 |\r\n| ⭐⭐⭐ 进阶 | L2 | 单一产品，3-5个知识点，需组合分析 |\r\n| ⭐⭐⭐⭐ 高阶 | L2-L3 | 多产品组合，竞争替换，高净值客户 |\r\n| ⭐⭐⭐⭐⭐ 专家 | L3 | 综合方案，税务筹划，财富传承 |\r\n\r\n#### Question Bank Generation Prompt:\r\n\r\n```\r\nBased on the product profile provided, generate a question bank with:\r\n\r\n1. For each difficulty tier (基础/入门/进阶/高阶/专家):\r\n - 5 multiple choice questions (产品知识)\r\n - 3 case analysis questions\r\n - 3 objection handling scenarios\r\n - 2 competitive comparison questions\r\n - 1 closing technique exercise\r\n\r\n2. Total: 65+ questions per product\r\n\r\n3. For each question, provide:\r\n - Question text\r\n - Difficulty level (1-5)\r\n - Category (产品知识/异议处理/案例分析/竞品对比/促成话术)\r\n - Ideal answer / model response\r\n - Evaluation criteria (excellent/good/needs-improvement)\r\n - Coaching tips for the trainer\r\n```\r\n\r\n---\r\n\r\n### 4. Personalized Training Scheduler / 个性化训练调度引擎（方法论演示）\r\n\r\n> **⚠️ 教学演示**：以下调度算法、代理人画像及日程数据均为**教学方法论的概念性展示**。**本技能不实际采集、存储或处理任何代理人或客户数据**。所有姓名、日程、业绩数据均为虚构示例，仅用于说明逻辑框架。\r\n\r\n**Input factors:**\r\n\r\n```\r\nAgent Profile (Level + Weak Points)\r\n +\r\nToday's Client Schedule (Who → What need → What product)\r\n +\r\nProduct Priority Matrix\r\n =\r\nPersonalized Daily Training Plan\r\n```\r\n\r\n**Scheduling Algorithm:**\r\n\r\n```python\r\ndef generate_daily_training_plan(agent_profile, daily_schedule, products):\r\n \"\"\"\r\n Generate personalized training plan for the day.\r\n \"\"\"\r\n # Step 1: Identify today's client visit products\r\n today_products = extract_products_from_schedule(daily_schedule)\r\n \r\n # Step 2: Get agent's weakness areas for these products\r\n weakness_map = get_weakness_for_products(\r\n agent_profile.weak_points, \r\n today_products\r\n )\r\n \r\n # Step 3: Calculate training time available\r\n available_minutes = calculate_available_training_time(daily_schedule)\r\n \r\n # Step 4: Prioritize by impact × weakness × product value\r\n training_queue = prioritize_training(\r\n weakness_map,\r\n today_products,\r\n agent_profile.level,\r\n time_constraint=available_minutes\r\n )\r\n \r\n # Step 5: Generate session plan\r\n sessions = split_into_sessions(training_queue, available_minutes)\r\n \r\n return {\r\n \"date\": today,\r\n \"agent\": agent_profile.name,\r\n \"total_minutes\": available_minutes,\r\n \"sessions\": sessions,\r\n \"focus_products\": today_products,\r\n \"key_objectives\": get_key_objectives(training_queue)\r\n }\r\n```\r\n\r\n**Example Daily Training Plan:**\r\n\r\n```json\r\n{\r\n \"date\": \"2026-05-05\",\r\n \"agent\": \"张明\",\r\n \"level\": \"L2\",\r\n \"total_minutes\": 90,\r\n \"sessions\": [\r\n {\r\n \"time\": \"08:00-08:20\",\r\n \"duration\": 20,\r\n \"type\": \"晨间快练\",\r\n \"mode\": \"快问快答\",\r\n \"focus\": \"年金险产品知识（高频问题5题）\",\r\n \"product\": \"福享人生终身寿险\",\r\n \"objective\": \"巩固年金转换权的计算逻辑\"\r\n },\r\n {\r\n \"time\": \"12:30-13:00\",\r\n \"duration\": 30,\r\n \"type\": \"午间强化\",\r\n \"mode\": \"情景对练\",\r\n \"focus\": \"健康险异议处理\",\r\n \"scenario\": \"客户：\"我有社保，不需要商业医疗险\"\",\r\n \"product\": \"康健医疗保险\",\r\n \"level\": \"⭐⭐⭐ 进阶\",\r\n \"coaching_tips\": \"引导客户认识到社保报销比例上限，用自费药比例对比引发需求\"\r\n },\r\n {\r\n \"time\": \"17:30-18:30\",\r\n \"duration\": 40,\r\n \"type\": \"晚间复盘\",\r\n \"mode\": \"案例分析 + 角色扮演\",\r\n \"focus\": \"私企业主综合保障方案\",\r\n \"scenario\": \"45岁私企老板，年收入200万，已有多份保单，如何做加保方案？\",\r\n \"products\": [\"终身寿险+万能账户\", \"高端医疗\", \"企业财产险\"],\r\n \"level\": \"⭐⭐⭐⭐ 高阶\",\r\n \"model_response_guide\": \"从家庭资产与企业资产隔离角度切入，引出终身寿险的债务隔离和传承功能\"\r\n }\r\n ],\r\n \"key_metrics_to_track\": [\r\n \"异议处理响应时间（目标<30秒）\",\r\n \"产品知识点正确率（目标>85%）\",\r\n \"方案组合完整性（3单以上产品覆盖）\"\r\n ]\r\n}\r\n```\r\n\r\n---\r\n\r\n### 5. Interactive Training Session / 智能陪练对话引擎\r\n\r\n**Session modes:**\r\n\r\n| Mode | Description | Duration | Best For |\r\n|------|-------------|----------|----------|\r\n| **快问快答** | Rapid-fire Q&A | 5-10 min | Pre-meeting warmup |\r\n| **情景对练** | Role-play (client vs. agent) | 15-30 min | Skill practice |\r\n| **案例研讨** | Real case analysis | 20-40 min | Advanced agents |\r\n| **异议攻关** | Objection busting focus | 10-15 min | Weak point training |\r\n| **综合考核** | Full simulation exam | 30-60 min | Level assessment |\r\n\r\n**Real-time coaching during training:**\r\n\r\n```\r\nAgent Response\r\n │\r\n ▼\r\n[Natural Language Understanding] → Extract key claims, tone, strategy\r\n │\r\n ▼\r\n[Evaluation Engine]\r\n ├─ Product knowledge accuracy ✓/✗\r\n ├─ Objection handling effectiveness (1-5)\r\n ├─ Compliance adherence ✓/✗\r\n ├─ Closing attempt timing (good/early/late/missing)\r\n ├─ Client empathy signals ✓/✗\r\n └─ Product combination logic ✓/✗\r\n │\r\n ▼\r\n[Real-time Coaching Feedback]\r\n ├─ Immediate tip (if struggling): \"💡 提示：可以先问客户目前的保障缺口...\"\r\n ├─ Completion praise (if excellent): \"🌟 完美！您已经很好地识别了客户需求\"\r\n └─ Post-question summary: \"本轮得分 85/100。建议加强：竞品对比环节\"\r\n```\r\n\r\n**Training session flow:**\r\n\r\n```\r\n1. 导入 (5%) → 介绍训练目标和产品背景\r\n2. 暖场 (10%) → 快问快答热身，激活产品知识\r\n3. 主体 (60%) → 情景对练：客户角色扮演 + 实时点评\r\n4. 复盘 (20%) → AI给出详细反馈：优点/不足/改进建议\r\n5. 行动 (5%) → 下次拜访的具体行动计划\r\n```\r\n\r\n---\r\n\r\n### 6. Effect Assessment & Progress Tracking / 效果评估与进度追踪\r\n\r\n**Metrics tracked per session:**\r\n\r\n| Metric | Definition | Target |\r\n|--------|------------|--------|\r\n| **产品知识得分** | 知识点正确率 | L1: ≥70%, L2: ≥80%, L3: ≥90% |\r\n| **异议处理时效** | 从异议提出到满意回答的时间 | < 30秒 |\r\n| **促成成功率** | 能否自然引入促成信号 | ≥ 1次有效尝试 |\r\n| **话术合规率** | 合规敏感词使用正确性 | 100% |\r\n| **方案完整性** | 保障覆盖广度 | ≥ 3个维度 |\r\n\r\n**Progress report structure:**\r\n\r\n```markdown\r\n## 📊 代理人张明 训练报告 - 2026-05-05\r\n\r\n### 综合得分: ⭐⭐⭐⭐ (78/100)\r\n\r\n| 维度 | 本次得分 | 较上次 | 目标 |\r\n|------|---------|--------|------|\r\n| 产品知识 | 82/100 | ↑5 | 80+ |\r\n| 异议处理 | 71/100 | ↓3 | 75+ |\r\n| 促成技巧 | 85/100 | ↑8 | 80+ |\r\n| 合规话术 | 95/100 | →0 | 100 |\r\n| 方案设计 | 72/100 | ↑12 | 75+ |\r\n\r\n### 🔥 本次表现亮点\r\n1. 养老规划方案逻辑清晰，能结合客户生命周期讲解\r\n2. 合规话术使用规范，犹豫期/退保说明完整\r\n\r\n### ⚠️ 需要加强\r\n1. 健康险异议处理：回应\"已有社保\"时过于被动，应主动算账\r\n2. 竞品对比：对中国平安主要产品线不够熟悉\r\n\r\n### 📅 明日训练重点\r\n- 产品：康健医疗保险（健康告知流程）\r\n- 场景：竞品替换（平安福 vs. XX福）\r\n- 时长：30分钟情景对练 + 10分钟快问快答\r\n```\r\n\r\n---\r\n\r\n## Workflow / 标准工作流程\r\n\r\n> **⚠️ 重要提示**：以下工作流展示的是**培训场景的教学参考**。所有销售话术和异议处理内容均为培训素材，实际销售行为须遵循《保险法》及相关监管规定，经持牌保险专业人士审核。\r\n\r\n### Mode 1: Quick Start (已知产品 + 快速训练)\r\n\r\n```\r\nUser: \"帮我准备明天拜访客户B的训练，他是私企老板，对健康险感兴趣\"\r\n │\r\n ▼\r\n[Step 1] 获取代理人信息 → 张明，L2，弱项：健康险异议处理\r\n[Step 2] 识别拜访产品 → 康健医疗保险（目标：替换平安福）\r\n[Step 3] 生成训练计划 → 午间30分钟：健康险异议处理对练\r\n[Step 4] 开始陪练 → 情景对练：私企业主健康险需求挖掘\r\n[Step 5] 实时反馈 → 异议处理评分：71/100，给出改进建议\r\n[Step 6] 报告输出 → 训练报告 + 明日拜访话术优化建议\r\n```\r\n\r\n### Mode 2: Product Document Upload (上传产品文档)\r\n\r\n```\r\nUser: [上传 XX保险公司福享人生终身寿险 产品手册 PDF]\r\n │\r\n ▼\r\n[Step 1] 解析文档 → 提取产品结构、条款、卖点\r\n[Step 2] 生成产品画像 → Structured JSON Profile\r\n[Step 3] 生成问题库 → 65+道题目（5难度×8类别）\r\n[Step 4] 生成参考题库 → 作为AI对话上下文，**不持久存储**\r\n[Step 5] 等待选择 → \"请选择训练模式：快问快答 / 情景对练 / 案例研讨\"\r\n```\r\n\r\n### Mode 3: Full Agent Assessment (全面能力评估)\r\n\r\n```\r\nUser: \"帮我评估代理人李华的综合能力，她入职8个月，主要卖重疾险\"\r\n │\r\n ▼\r\n[Step 1] 建立代理人档案 → L1入门级，8个月，重疾险方向\r\n[Step 2] 产品文档上传 → 重疾险产品手册\r\n[Step 3] 综合考核 → 30题产品知识 + 5个情景对练\r\n[Step 4] 生成能力雷达图 → 6维度能力可视化\r\n[Step 5] 制定成长路径 → 90天训练计划\r\n```\r\n\r\n---\r\n\r\n## Input / Output Specifications / 输入输出规范\r\n\r\n### Input\r\n\r\n| Input Type | Description | Example |\r\n|------------|-------------|---------|\r\n| 代理人档案 | JSON/文本描述 | 姓名、级别、工龄、业绩、弱项 |\r\n| 产品文档 | PDF/Word/TXT/图片 | 保险产品手册、条款、计划书 |\r\n| 当日行程 | 文本/日历 | 09:00晨会 / 10:30拜访客户A |\r\n| 训练指令 | 自然语言 | \"帮我准备健康险的陪练\" |\r\n| 客户信息 | 文本描述 | \"45岁私企老板，年收入200万\" |\r\n\r\n### Output\r\n\r\n| Output Type | Description |\r\n|-------------|-------------|\r\n| 产品画像JSON | 结构化产品信息 |\r\n| 问题库 | 65+道分类分级题目 |\r\n| 训练计划 | 分钟级个性化日程 |\r\n| 陪练对话 | 实时AI角色扮演 |\r\n| 评估报告 | 评分 + 改进建议 + 雷达图 |\r\n| 成长路径 | 30/60/90天训练建议 |\r\n\r\n---\r\n\r\n## Integration Notes / 集成说明\r\n\r\n**Data privacy:**\r\n- All agent and client data remains local / within the company's system\r\n- No sensitive PII should be included in training documents\r\n- Comply with China CBIRC insurance sales compliance regulations\r\n\r\n**Lianxi with other Skills:**\r\n- `insurance-bidding-pro`: Use product analysis for bidding scenarios\r\n- `insurance-private-domain-ops`: Link training completion to customer follow-up\r\n- `insurance-claims-intelligence`: Train agents on claim processes for better client communication\r\n\r\n---\r\n\r\n## Disclaimer / 免责声明\r\n\r\n> ⚠️ **Training is advisory only.** This skill provides coaching materials, question banks,\r\n> and simulation training for insurance agent development. All final sales advice,\r\n> compliance decisions, and product recommendations must be reviewed by licensed\r\n> insurance professionals and comply with CBIRC regulations. Model answers represent\r\n> reference best practices, not guaranteed outcomes.\n\nFile v5.2.2:README.md\n\n# Insurance Agent Intelligent Trainer / 保险代理人智能陪练系统\r\n\r\n> **English** — AI-powered insurance agent coaching platform. Auto-parses product documents,\r\n> generates question banks, assesses agent competency (L1/L2/L3), schedules personalized daily\r\n> training based on client visits, and runs interactive role-play drills. Benchmarked against\r\n> AIA, Ping An, and Alibaba Cloud insurance training systems.\r\n\r\n> **中文** — 保险代理人智能陪练系统。自动解析产品文档、生成问题库、评估代理人能力等级、\r\n> 结合当日客户拜访行程安排个性化训练、进行情景对练。对标友邦保险、平安保险、阿里云智能陪练水平。\r\n\r\n---\r\n\r\n## ✨ Features / 核心功能\r\n\r\n### 🚀 Product Document Parser / 产品文档解析\r\n- **Input formats**: PDF, Word (.docx), scanned images (OCR), plain text\r\n- **Output**: Structured JSON product profile with coverage, terms, selling points, exclusions\r\n- **Official data**: Integrates with China Welfare Lottery & Sports Lottery public APIs for training case design\r\n\r\n### 🎯 Question Bank Generator / 问题库自动生成\r\n- **116+ questions** per product across **8 categories** × **5 difficulty tiers**\r\n- Categories: Product Knowledge · Objection Handling · Case Analysis · Competitive Comparison · Closing Techniques · Compliance Scripts · Needs Discovery · Digital Planning\r\n- Difficulty: ⭐基础 → ⭐⭐⭐⭐⭐专家 (L1–L3 agents)\r\n\r\n### 👤 Agent Profiling & Assessment / 代理人画像与评估\r\n- **3-tier competency model**: L1 (Beginner) / L2 (Intermediate) / L3 (Advanced)\r\n- **6-dimension radar chart**: Product Knowledge · Needs Discovery · Objection Handling · Closing · Compliance · Customer Relations\r\n- **Growth roadmap**: 30/60/90-day personalized development plans\r\n\r\n### 📅 Personalized Daily Training Scheduler / 个性化训练调度\r\n- Analyzes agent's daily client visit schedule\r\n- Maps visit products → training focus areas\r\n- Generates minute-level daily training plan with session recommendations\r\n\r\n### 🗣️ Interactive Training Modes / 智能陪练模式\r\n- **快问快答** — Rapid-fire Q&A warmup (5–10 min)\r\n- **情景对练** — Role-play (15–30 min)\r\n- **案例研讨** — Case analysis (20–40 min)\r\n- **异议攻关** — Objection busting focus (10–15 min)\r\n- **综合考核** — Full simulation exam (30–60 min)\r\n\r\n### 📊 Real-time Effect Tracking / 效果实时追踪\r\n- Session-level scoring (6 dimensions, 100-point scale)\r\n- 30-day trend analysis with improvement indicators\r\n- Radar chart visualization of competency progress\r\n\r\n---\r\n\r\n## 🚀 Quick Start\r\n\r\n### Installation / 安装\r\n\r\n```bash\r\n# Install via ClawHub\r\nopenclaw skills install insurance-agent-trainer\r\n\r\n# Or via npm\r\nnpx clawhub install @gechengling/insurance-agent-trainer\r\n```\r\n\r\n### Basic Usage / 基本使用\r\n\r\n#### 1. Upload product document and generate question bank\r\n```markdown\r\nUser: 请帮我解析[产品名称]的产品文档，并生成L2级别的问题库\r\n→ AI: 解析文档 → 生成116道题目 → 输出问题库JSON + Markdown报告\r\n```\r\n\r\n#### 2. Create agent profile and get daily training plan\r\n```markdown\r\nUser: 帮我安排代理人张明今天的训练计划，他今天要拜访3个客户（健康险+养老+教育金）\r\n→ AI: 分析行程 → 评估弱项 → 生成3段训练（共90分钟）\r\n```\r\n\r\n#### 3. Start interactive training session\r\n```markdown\r\nUser: 开始健康险的异议处理对练，我是L2级别\r\n→ AI: 启动情景对练 → AI扮演客户 → 实时点评 → 训练报告\r\n```\r\n\r\n#### 4. Full agent competency assessment\r\n```markdown\r\nUser: 帮我评估代理人李华的综合能力，她入职8个月，主攻重疾险\r\n→ AI: 全面评估 → 能力雷达图 → 90天成长路径\r\n```\r\n\r\n---\r\n\r\n## 📁 File Structure / 文件结构\r\n\r\n```\r\ninsurance-agent-trainer/\r\n├── SKILL.md # 主技能文件（含完整工作流程）\r\n├── README.md # 本文件（双语）\r\n├── references/\r\n│ ├── question_bank_templates.md # 问题库模板（8类别×5难度）\r\n│ ├── agent_profile_template.md # 代理人画像模板 + 成长路径\r\n│ └── training_evaluation_rubric.md # 训练效果评估量表\r\n└── scripts/\r\n ├── product_parser.py # 产品文档解析 + 彩票数据获取\r\n ├── question_generator.py # 问题库自动生成器\r\n └── training_scheduler.py # 训练日程调度器\r\n```\r\n\r\n---\r\n\r\n## 📋 Script Usage / 脚本使用\r\n\r\n```python\r\n# 1. 产品文档解析\r\nfrom scripts.product_parser import parse_insurance_product_from_text\r\n\r\ntext = open(\"product_manual.txt\").read()\r\nprofile = parse_insurance_product_from_text(text)\r\nprint(profile)\r\n\r\n# 2. 官方数据获取（供训练案例使用）\r\nfrom scripts.product_parser import get_training_case_data\r\nssq_data = get_training_case_data(\"ssq\", limit=10) # 双色球\r\ndlt_data = get_training_case_data(\"dlt\", limit=10) # 大乐透\r\n\r\n# 3. 生成问题库\r\nfrom scripts.question_generator import generate_question_bank\r\nqb = generate_question_bank(profile, agent_level=\"L2\", questions_per_category=5)\r\nprint(f\"生成 {qb['meta']['total_questions']} 道题目\")\r\n\r\n# 4. 生成每日训练计划\r\nfrom scripts.training_scheduler import generate_daily_training_plan, format_training_plan_markdown\r\nplan = generate_daily_training_plan(agent_profile, daily_schedule, [\"健康险\", \"年金险\"])\r\nprint(format_training_plan_markdown(plan))\r\n```\r\n\r\n---\r\n\r\n## 🔗 Related Skills / 相关技能\r\n\r\n| Skill | Relationship |\r\n|-------|-------------|\r\n| `insurance-bidding-pro` | Use product analysis for corporate insurance bidding |\r\n| `insurance-private-domain-ops` | Link training completion to customer retention campaigns |\r\n| `insurance-claims-intelligence` | Train agents on claims processes for better client communication |\r\n| `insurance-actuarial-cn` | Deep actuarial knowledge for advanced product training |\r\n\r\n---\r\n\r\n## ⚠️ Disclaimer / 免责声明\r\n\r\n> This skill provides coaching materials, question banks, and simulation training for\r\n> insurance agent development. All final sales advice, compliance decisions, and product\r\n> recommendations must be reviewed by licensed insurance professionals in compliance with\r\n> China CBIRC regulations. Model answers represent reference best practices, not guaranteed outcomes.\r\n\r\n---\r\n\r\n## 📄 License / 许可证\r\n\r\nMIT-0 — Free for commercial use, no attribution required.\r\n\r\n---\r\n\r\n*Built for China life insurance companies, insurtech platforms, and agency leaders.*\r\n*为中国寿险公司、保险科技平台、营业部主管打造。*\n\nFile v5.2.2:_meta.json\n\n{\n \"ownerId\": \"kn74e704j3ygjcygnpf02rdvd185js13\",\n \"slug\": \"insurance-agent-trainer\",\n \"version\": \"5.2.2\",\n \"publishedAt\": 1782652458763\n}\n\nFile v5.2.2:references/agent_profile_template.md\n\n# Agent Profile & Skill Assessment Framework / 代理人画像与能力评估体系\r\n\r\n## 一、代理人画像标准模板 / Agent Profile Standard Template\r\n\r\n```json\r\n{\r\n \"agent_id\": \"AG_XXXX_XXXX\",\r\n \"basic_info\": {\r\n \"name\": \"[化名，保护隐私]\",\r\n \"gender\": \"男/女\",\r\n \"age\": 32,\r\n \"education\": \"本科\",\r\n \"certifications\": [\r\n \"保险代理人资格证书（必选）\",\r\n \"健康险销售资质（可选）\",\r\n \"理财规划师（可选）\",\r\n \"CFP国际金融理财师（可选）\"\r\n ],\r\n \"entry_date\": \"2024-03-01\",\r\n \"tenure_months\": 14,\r\n \"agency\": \"[营业部和团队名称]\"\r\n },\r\n \"skill_level\": {\r\n \"current\": \"L2\",\r\n \"label\": \"进阶级\",\r\n \"certified_date\": \"2025-06-01\",\r\n \"last_assessment_date\": \"2026-03-15\",\r\n \"target_level\": \"L3\",\r\n \"target_date\": \"2026-12-31\"\r\n },\r\n \"performance_metrics\": {\r\n \"quarterly\": {\r\n \"premium_target\": 150000,\r\n \"premium_actual\": 128000,\r\n \"target_achievement_rate\": 0.853,\r\n \"policy_count\": 8,\r\n \"avg_policy_size\": 16000,\r\n \"new_customers\": 5,\r\n \"existing_customers\": 3,\r\n \"closing_rate\": 0.32,\r\n \"visit_count\": 25\r\n },\r\n \"historical\": {\r\n \"months_1_3_premium\": 45000,\r\n \"months_4_6_premium\": 68000,\r\n \"months_7_12_premium\": 142000,\r\n \"months_13_plus_trend\": \"+12%\"\r\n }\r\n },\r\n \"product_mastery_radar\": {\r\n \"term_life\": {\r\n \"score\": 0.85,\r\n \"label\": \"熟练\",\r\n \"details\": \"条款清晰，能独立完成产品对比\"\r\n },\r\n \"whole_life\": {\r\n \"score\": 0.72,\r\n \"label\": \"良好\",\r\n \"details\": \"万能账户理解较好，传承功能讲解有提升空间\"\r\n },\r\n \"critical_illness\": {\r\n \"score\": 0.58,\r\n \"label\": \"薄弱\",\r\n \"details\": \"病种定义理解不足，脑中风后遗症与同类产品对比不熟悉\"\r\n },\r\n \"medical_insurance\": {\r\n \"score\": 0.80,\r\n \"label\": \"熟练\",\r\n \"details\": \"社保对比算账能力强\"\r\n },\r\n \"annuity\": {\r\n \"score\": 0.45,\r\n \"label\": \"薄弱\",\r\n \"details\": \"IRR计算不熟练，年金转换逻辑讲解不够清晰\"\r\n },\r\n \"investment_linked\": {\r\n \"score\": 0.38,\r\n \"label\": \"薄弱\",\r\n \"details\": \"净值波动解释不够形象，高端客户需求识别不足\"\r\n }\r\n },\r\n \"competency_assessment\": {\r\n \"needs_identification\": {\r\n \"score\": 0.75,\r\n \"label\": \"良好\",\r\n \"evidence\": \"能通过提问识别客户基础需求\"\r\n },\r\n \"product_presentation\": {\r\n \"score\": 0.68,\r\n \"label\": \"及格\",\r\n \"evidence\": \"讲解清晰但缺乏个性化，客户参与度一般\"\r\n },\r\n \"objection_handling\": {\r\n \"score\": 0.62,\r\n \"label\": \"薄弱\",\r\n \"evidence\": \"异议处理反应较慢，算账说服力不足\"\r\n },\r\n \"closing_technique\": {\r\n \"score\": 0.78,\r\n \"label\": \"良好\",\r\n \"evidence\": \"促成时机把握较好，促成后跟单能力强\"\r\n },\r\n \"compliance_awareness\": {\r\n \"score\": 0.92,\r\n \"label\": \"优秀\",\r\n \"evidence\": \"双录操作规范，合规话术使用到位\"\r\n },\r\n \"customer_relationship\": {\r\n \"score\": 0.88,\r\n \"label\": \"优秀\",\r\n \"evidence\": \"老客户复购率高，转介绍渠道稳定\"\r\n }\r\n },\r\n \"training_history\": [\r\n {\r\n \"date\": \"2026-03-15\",\r\n \"product\": \"健康险\",\r\n \"type\": \"情景对练\",\r\n \"duration_minutes\": 30,\r\n \"score_before\": 65,\r\n \"score_after\": 78,\r\n \"improvement\": \"+13\"\r\n },\r\n {\r\n \"date\": \"2026-04-01\",\r\n \"product\": \"终身寿险\",\r\n \"type\": \"案例研讨\",\r\n \"duration_minutes\": 45,\r\n \"score_before\": 70,\r\n \"score_after\": 74,\r\n \"improvement\": \"+4\"\r\n }\r\n ],\r\n \"personalized_weak_points\": [\r\n \"健康险条款定义讲解不够通俗，需要案例化表达\",\r\n \"IRR计算不够熟练，高净值客户对收益率敏感时容易卡壳\",\r\n \"竞品对比知识储备不足，主要竞品（平安、太平洋）产品线不熟悉\"\r\n ],\r\n \"personalized_strong_points\": [\r\n \"缘故市场开发能力强，信任背书效果好\",\r\n \"老客户维护细致，复购率高于团队平均\",\r\n \"合规意识强，从未发生投诉\"\r\n ],\r\n \"daily_schedule_sample\": {\r\n \"workday_pattern\": \"标准工作日（09:00-18:00）\",\r\n \"meeting_hours\": \"09:00-09:30（晨会）\",\r\n \"avg_visits_per_day\": 3.5,\r\n \"travel_pattern\": \"以市区客户拜访为主，偶尔郊区\"\r\n }\r\n}\r\n```\r\n\r\n---\r\n\r\n## 二、能力评估量表 / Competency Assessment Rubric\r\n\r\n### L1 → L2 晋升评估标准\r\n\r\n| 能力维度 | L1入门标准 | L2进阶标准 | L3专家标准 |\r\n|---------|-----------|-----------|-----------|\r\n| **产品知识** | 能复述产品基本信息（名称、保障、缴费期）| 能对比3款以上同类产品，能解释条款差异 | 能做竞品逐条对比，能处理复杂条款解释 |\r\n| **需求挖掘** | 能问出客户\"要买什么类型\" | 能通过追问发现客户\"真实担心和期望\" | 能挖掘高净值客户资产隔离/传承/税务需求 |\r\n| **异议处理** | 能处理2-3个高频异议（有固定话术）| 能灵活运用算账、类比、共情等方法 | 能反客为主，将异议转化为成交契机 |\r\n| **促成技巧** | 有促成意识，能做1次标准促成 | 能识别多个促成信号，能做2-3次渐进式促成 | 促成无痕自然，让客户感觉是自己在决定买 |\r\n| **合规意识** | 双录操作规范 | 合规话术灵活运用，能识别客户误导风险 | 能主动向团队新人传授合规经验 |\r\n| **客户经营** | 能维护老客户，基础问候 | 能做老客户加保需求激活，转介绍开发 | 能建立高净值客户长期顾问关系，跨品类经营 |\r\n| **数字能力** | 保额保费简单计算 | IRR、内部收益率、节税计算熟练 | 能做综合财务规划方案（保险+理财+税务） |\r\n\r\n---\r\n\r\n## 三、能力雷达图数据 / Competency Radar Chart Data\r\n\r\n```python\r\ndef get_radar_chart_data(agent_profile):\r\n \"\"\"\r\n Generate radar chart data for agent competency visualization.\r\n Each dimension scores 0-100.\r\n \"\"\"\r\n return {\r\n \"labels\": [\r\n \"产品知识\",\r\n \"需求挖掘\",\r\n \"异议处理\",\r\n \"促成技巧\",\r\n \"合规意识\",\r\n \"客户经营\",\r\n \"数字规划\"\r\n ],\r\n \"L1_threshold\": [70, 60, 50, 55, 80, 65, 40],\r\n \"L2_threshold\": [80, 75, 70, 75, 90, 80, 65],\r\n \"L3_threshold\": [90, 88, 85, 88, 95, 90, 80],\r\n \"current_scores\": [\r\n int(agent_profile[\"competency_assessment\"][\"product_knowledge\"][\"score\"] * 100),\r\n int(agent_profile[\"competency_assessment\"][\"needs_identification\"][\"score\"] * 100),\r\n int(agent_profile[\"competency_assessment\"][\"objection_handling\"][\"score\"] * 100),\r\n int(agent_profile[\"competency_assessment\"][\"closing_technique\"][\"score\"] * 100),\r\n int(agent_profile[\"competency_assessment\"][\"compliance_awareness\"][\"score\"] * 100),\r\n int(agent_profile[\"competency_assessment\"][\"customer_relationship\"][\"score\"] * 100),\r\n 65 # 数字规划维度（需单独评估）\r\n ]\r\n }\r\n```\r\n\r\n---\r\n\r\n## 四、成长路径模板 / Development Roadmap Template\r\n\r\n```markdown\r\n# 代理人成长路径 / Agent Development Roadmap\r\n\r\n## 张明 — L2 → L3 专家级成长计划（90天）\r\n\r\n### 当前状态（基线）\r\n- 综合得分：72/100\r\n- 主要短板：健康险异议处理（58分）、IRR计算（45分）\r\n- 主要优势：客户关系（88分）、促成技巧（78分）\r\n\r\n### 30天目标（第一阶段）\r\n**主题**：健康险专项突破\r\n\r\n| 周次 | 训练内容 | 训练量 | KPI目标 |\r\n|------|---------|--------|--------|\r\n| 第1周 | 健康险产品知识巩固（每日10题） | 50题 | 正确率≥75% |\r\n| 第2周 | 健康险异议处理情景对练 | 5个场景 | 时效<45秒 |\r\n| 第3周 | 健康险算账练习（社保管控vs商业险） | 10个案例 | 能独立算出差异 |\r\n| 第4周 | 综合演练：健康险完整销售流程 | 3次完整陪练 | 综合得分≥72 |\r\n\r\n### 60天目标（第二阶段）\r\n**主题**：高净值客户专项 + IRR计算\r\n\r\n| 周次 | 训练内容 | 训练量 | KPI目标 |\r\n|------|---------|--------|--------|\r\n| 第5周 | IRR计算原理与演示练习 | 20个算例 | 3分钟内完成计算 |\r\n| 第6周 | 年金险IRR对比（vs定期存款/国债） | 10个场景 | 能清晰讲解 |\r\n| 第7周 | 高净值客户情景对练（私企业主） | 5个场景 | 方案设计≥3产品 |\r\n| 第8周 | 竞品对比（平安/太平洋/新华/泰康）| 各产品线1份对比表 | 熟练程度≥80% |\r\n\r\n### 90天目标（第三阶段）\r\n**主题**：综合实战 + 晋升评估\r\n\r\n| 周次 | 训练内容 | 训练量 | KPI目标 |\r\n|------|---------|--------|--------|\r\n| 第9周 | 综合案例分析（家庭综合保障方案）| 10个案例 | 方案完整度≥85% |\r\n| 第10周 | 导师带教（指导1名L1新人）| 2次实战带教 | 教学评分≥80 |\r\n| 第11周 | 模拟晋升评估 | 3套综合试卷 | 总分≥85 |\r\n| 第12周 | **正式晋升评估** | 实战考核 | 达标则晋升L3 |\r\n\r\n### 成功标准\r\n| 指标 | 当前值 | 30天目标 | 60天目标 | 90天目标 |\r\n|------|--------|---------|---------|---------|\r\n| 综合得分 | 72 | 75 | 80 | 85 |\r\n| 健康险异议处理 | 58 | 70 | 78 | 82 |\r\n| IRR计算 | 45 | 60 | 78 | 85 |\r\n| 竞品对比 | 55 | 65 | 80 | 85 |\r\n| 促成综合得分 | 78 | 80 | 82 | 85 |\r\n```\n\nFile v5.2.2:references/question_bank_templates.md\n\n# Insurance Agent Training Question Bank Templates / 保险代理人训练问题库模板\r\n\r\n## 模板说明 / Template Description\r\n\r\n本文档提供标准化的问题库模板，支持从产品文档自动生成题库。\r\n每类问题均按5级难度梯度设计，适用于L1-L3各级别代理人。\r\n\r\n---\r\n\r\n## 一、产品知识题 / Product Knowledge Questions\r\n\r\n### 1.1 选择题模板 (Multiple Choice)\r\n\r\n```markdown\r\n## 【单选题】⭐ 基础题\r\n\r\n**题目**: [产品名称]的[核心条款/数值/规则]是？\r\n\r\nA. [选项A - 错误]\r\nB. [选项B - 正确]\r\nC. [选项C - 干扰项]\r\nD. [选项D - 混淆项]\r\n\r\n**正确答案**: B\r\n\r\n**知识点**: [章节名，例如：等待期条款]\r\n\r\n**适用级别**: L1新人\r\n\r\n**参考话术**:\r\n> \"这款产品的等待期是90天，这也是行业大多数重疾险的标准等待期。在等待期内发生重疾，我们退还已交保费，但不会赔付保险金。不过，意外伤害导致的重大疾病是没有等待期限制的，这一点对客户很重要。\"\r\n\r\n---\r\n\r\n## 【多选题】⭐⭐⭐ 进阶题\r\n\r\n**题目**: 满足以下哪些条件时，[产品名称]可以申请[保险金类型]？\r\n\r\nA. [条件A]\r\nB. [条件B]\r\nC. [条件C]\r\nD. [条件D]\r\n\r\n**正确答案**: [A,C]\r\n\r\n**知识点**: [条款名]\r\n\r\n**适用级别**: L2进阶\r\n\r\n**解析**:\r\n- A正确，因为...\r\n- B错误，因为...\r\n- C正确，因为...\r\n- D错误，因为...\r\n\r\n**评估标准**:\r\n- 优秀：能准确识别全部正确选项，并解释原因\r\n- 合格：选择正确，但解释不完整\r\n- 不合格：漏选或错选关键选项\r\n```\r\n\r\n### 1.2 计算题模板\r\n\r\n```markdown\r\n## 【计算题】⭐⭐⭐⭐ 高阶题\r\n\r\n**场景**: [年龄]岁，[职业类型]，[年收入]，现有[保障情况]，希望规划[财务目标]\r\n\r\n**计算任务**:\r\n1. 计算该客户的合理保额区间\r\n2. 计算10年缴/20年缴的年缴保费\r\n3. 对比不同缴费期的总保费差异\r\n\r\n**参考计算过程**:\r\n\r\n| 指标 | 计算公式 | 结果 |\r\n|------|---------|------|\r\n| 生命价值法保额 | 年收入×(65-当前年龄)×10% | XX万 |\r\n| 家庭需求法保额 | 家庭负债+子女教育+养老预留-已有资产 | XX万 |\r\n| 合理保费(10年缴) | 家庭年收入×15% | XX元/年 |\r\n| 合理保费(20年缴) | 家庭年收入×15% | XX元/年 |\r\n\r\n**适用级别**: L2-L3\r\n\r\n**评估标准**:\r\n- 优秀：两种方法均正确计算，并给出综合建议\r\n- 合格：至少一种方法正确，有基本逻辑\r\n- 不合格：计算错误或逻辑混乱\r\n```\r\n\r\n---\r\n\r\n## 二、异议处理题 / Objection Handling Questions\r\n\r\n### 异议处理话术框架 (A-C-E Framework)\r\n\r\n```\r\nA - Acknowledge（认同情绪）: \"是的，我理解您的顾虑...\"\r\nC - Clarify（澄清需求）: \"您主要担心的是...对吗？\"\r\nE - Educate（教育引导）: \"其实，根据您的实际情况...\"\r\n```\r\n\r\n### 2.1 高频异议处理模板\r\n\r\n```markdown\r\n## 【异议处理】异议类型：需求认知类\r\n\r\n**客户异议**: \"我有社保了，不需要商业保险。\"\r\n\r\n**适用产品**: 医疗险、重疾险\r\n\r\n**L1 参考话术（入门级）**:\r\n> \"李先生，社保确实是我们国家的基本保障，很重要。不过社保有一个局限，就是有报销比例和范围的限制。比如，进口药、自费药社保是不能报销的，重大疾病治疗中的很多靶向药一个月就要几万块，都是社保外费用。商业保险正好可以报销这部分。我帮您算一下，30岁男性，一年500元的医疗险，社保报销后超过1万的部分可以100%报销，包括社保外的自费药。这个您看对您有没有用？\"\r\n\r\n**L2 参考话术（进阶级）**:\r\n> \"李先生，我先问您一个问题——如果您或家人突然生了一场大病，需要50万的治疗费用，社保能报销多少？\r\n> 实际上，社保平均报销比例大约在50%-60%左右，还有很多特效药、进口器材社保完全不报。50万的治疗费，社保大概只能报20-30万，还有20-30万的缺口需要自己承担。\r\n> 您年收入30万，拿出500元/年买个医疗险，撬动200万的保障，把这个风险缺口完全覆盖住。这是'小钱撬大保障'的逻辑，您觉得值不值？\"\r\n\r\n**L3 参考话术（专家级）**:\r\n> \"李先生，我们做个测算——\r\n> 假设您45岁，标准治疗一场癌症需要30-50万，社保报销约15-25万，自费部分约15-25万。\r\n> 更重要的是：大病治疗期间您可能3-5年无法正常工作，家庭收入中断，但房贷、孩子教育、父母赡养的费用不会停。\r\n> 一个人倒下，最危险的不是医疗费，而是'收入损失'和'家庭责任'。所以我建议您考虑一个'医疗险+重疾险'的组合——医疗险报销治疗费用，重疾险一次性给付，补偿收入损失。\r\n> 您目前的家庭责任有多重？\"\r\n\r\n**评估标准**:\r\n- 🌟 优秀：主动算账 + 量化风险 + 逻辑递进 + 引入重疾险组合\r\n- ✓ 合格：能反驳异议，有基本逻辑，但说服力不足\r\n- ✗ 不合格：直接否定客户，或未能给出合理解释\r\n```\r\n\r\n```markdown\r\n## 【异议处理】异议类型：产品比较类\r\n\r\n**客户异议**: \"XX公司的产品比你们的好，保费还便宜。\"\r\n\r\n**适用产品**: 全品类\r\n\r\n**L2 参考话术**:\r\n> \"张先生，您提到的XX公司确实是大公司，产品也不错。买东西货比三家是对的。\r\n> 不过保险产品比较有个特点——不能只看保费，要看'性价比'和'适合度'。\r\n> 我帮您做个对比表：[产品名称] vs [竞品]\"\r\n\r\n| 对比维度 | [产品名称] | [竞品] |\r\n|---------|-----------|--------|\r\n| 核心保障 | [我方优势] | [对方描述] |\r\n| 保障范围 | [详细] | [描述] |\r\n| 等待期 | 90天 | 180天 |\r\n| 特色功能 | [独有] | [缺失] |\r\n| 综合性价比 | ⭐⭐⭐⭐⭐ | ⭐⭐⭐ |\r\n\r\n> 选保险更重要的是选'条款'，不是选'公司'。因为保险公司受银保监会严格监管，每家都有实力赔付。\r\n> 您方便把他们的条款发给我吗？我帮您逐条对比，找到最适合您需求的方案。\r\n```\r\n\r\n```markdown\r\n## 【异议处理】异议类型：信任类\r\n\r\n**客户异议**: \"保险都是骗人的，到时候这也不赔那也不赔。\"\r\n\r\n**L1 参考话术**:\r\n> \"李先生，我能理解您的担心。这种印象主要来自于两个方面：一是早期保险销售不专业，导致客户买错了产品；二是很多人没有认真看过保险条款，不知道什么赔什么不赔。\r\n> 现在监管要求'双录'，保险合同也越来越透明。我会帮您把条款逐条解释清楚，您签字之前完全理解保障内容。我们公司有'秒赔'服务，只要是条款内的责任，上传资料后24小时内到账。\r\n> 您之前遇到的是什么情况？我帮您分析一下，是产品选择问题还是条款理解问题。\"\r\n\r\n**评估标准**:\r\n- 🌟 优秀：不反驳客户情绪，先共情，再引导，最后主动提出解决方案\r\n- ✓ 合格：能共情但解决方案模糊\r\n- ✗ 不合格：直接否定客户，或过度承诺\r\n```\r\n\r\n---\r\n\r\n## 三、案例分析题 / Case Analysis Questions\r\n\r\n### 3.1 标准案例模板\r\n\r\n```markdown\r\n## 【案例分析】⭐⭐⭐ 进阶案例\r\n\r\n### 客户档案\r\n| 项目 | 内容 |\r\n|------|------|\r\n| 姓名 | [化名] |\r\n| 年龄/性别 | 40岁男性 |\r\n| 职业 | 国企中层管理 |\r\n| 年收入 | 50万元 |\r\n| 婚姻状态 | 已婚，子女8岁 |\r\n| 现有保障 | 社保，房贷80万（剩余15年） |\r\n| 家庭压力 | 上有父母（60+岁），需赡养 |\r\n| 来访目的 | 想给孩子存教育金，顺便考虑养老 |\r\n\r\n### 客户需求分析（代理人填写）\r\n**核心需求**: [ ]\r\n**潜在需求**: [ ]\r\n**紧急程度**: [高/中/低]\r\n\r\n### 产品推荐方案（代理人设计）\r\n\r\n**推荐产品组合**:\r\n\r\n| 产品 | 险种 | 保障/功能 | 年缴保费 | 缴费期 |\r\n|------|------|---------|---------|--------|\r\n| 产品A | 终身寿险 | 身故保障+万能账户 | XX元 | X年 |\r\n| 产品B | 医疗险 | 覆盖社保内外医疗费 | XX元 | 1年 |\r\n| 产品C | 重疾险 | XX万重疾保障 | XX元 | X年 |\r\n| 产品D | 教育金 | 18岁起领取 | XX元 | X年 |\r\n\r\n**方案亮点**:\r\n1. [亮点1]\r\n2. [亮点2]\r\n3. [亮点3]\r\n\r\n**预估年缴总保费**: ___元（占年收入___%）\r\n\r\n### 评估标准\r\n\r\n| 评估维度 | 优秀 | 合格 | 不合格 |\r\n|---------|------|------|--------|\r\n| 需求挖掘 | 能发现养老+传承需求 | 聚焦教育金 | 仅推销产品 |\r\n| 产品组合 | 3单以上，覆盖全面 | 2单，基本合理 | 单品推销 |\r\n| 保费测算 | 合理，占比<20% | 基本合理 | 过高或过低 |\r\n| 逻辑表达 | 条理清晰，客户能听懂 | 有逻辑但不够流畅 | 逻辑混乱 |\r\n```\r\n\r\n### 3.2 高净值客户案例\r\n\r\n```markdown\r\n## 【案例分析】⭐⭐⭐⭐⭐ 专家级案例\r\n\r\n### 客户档案\r\n| 项目 | 内容 |\r\n|------|------|\r\n| 姓名 | [化名] |\r\n| 年龄/性别 | 50岁男性 |\r\n| 职业 | 私企老板（制造业） |\r\n| 年收入 | 200-500万元 |\r\n| 资产状况 | 净资产2000万+（含企业资产） |\r\n| 家庭状况 | 离异，子女2人（15岁、20岁） |\r\n| 风险担忧 | 企业债务风险、家庭传承、税务风险 |\r\n| 核心诉求 | 资产隔离 + 税务优化 + 子女传承 |\r\n\r\n### 需求挖掘问题链\r\n1. \"您企业目前负债情况如何？有没有个人连带担保？\"\r\n2. \"您希望孩子将来继承的是企业还是现金资产？\"\r\n3. \"您目前的税务顾问有没有做过个人所得税的专项规划？\"\r\n4. \"您对子女教育金的规划是希望确定性给付还是弹性安排？\"\r\n\r\n### 专家级方案设计要点\r\n\r\n**核心产品**: 终身寿险（高保额）+ 保险金信托\r\n\r\n| 方案层次 | 产品/工具 | 功能 | 核心价值 |\r\n|---------|---------|------|---------|\r\n| 第一层 | 高额终身寿险 | 身故保险金杠杆 | 10倍杠杆，指定传承 |\r\n| 第二层 | 保险金信托 | 信托架构 | 防止子女挥霍，按条件分批给付 |\r\n| 第三层 | 年金险 | 养老现金流 | 55岁起每年领取，保证品质养老 |\r\n| 第四层 | 高端医疗 | 医疗资源 | 全国三甲医院绿色通道，海外二次诊疗 |\r\n\r\n**关键销售逻辑**:\r\n> \"李总，您现在的企业资产和家庭资产没有做好隔离。一旦企业发生债务风险，家庭资产可能被追索。\r\n> 终身寿险的作用是：用5万/年的保费，撬动500万的身故保障，这500万是保险金，不属于遗产，不用于清偿债务，可以完整留给您的两个孩子。\r\n> 如果再通过保险金信托架构，可以设定孩子年满30岁才可领取，且每月领取不超过5万，防止他们一次性拿到大笔资金后挥霍。\r\n> 这是目前高净值客户最常用的'资产隔离+定向传承'的法律工具之一。\"\r\n\r\n**评估标准**:\r\n- 🌟 优秀：能识别企业债务隔离+传承核心需求，设计保险金信托架构，引入税务顾问联动\r\n- ✓ 合格：识别了传承需求，有基本方案但深度不足\r\n- ✗ 不合格：仍以推销产品为主，未触及核心风险\r\n```\r\n\r\n---\r\n\r\n## 四、促成话术题 / Closing Technique Questions\r\n\r\n### 4.1 促成信号识别\r\n\r\n```markdown\r\n## 【促成话术】促成信号识别与应对\r\n\r\n### 常见促成信号（出现即响应）\r\n1. 🔔 \"这个产品听起来不错，多少钱一年？\"\r\n2. 🔔 \"如果我现在买，什么时候生效？\"\r\n3. 🔔 \"能不能把保费降低一点？\"\r\n4. 🔔 \"我回去和家人商量一下。\"\r\n5. 🔔 \"我有个朋友也想买，能不能一起介绍？\"\r\n\r\n### L1 标准促成话术（假设成交法）\r\n> \"李先生，如果这款产品您觉得合适的话，我们今天就把计划书做出来，您看看每年存3万，存10年，保障终身，您看可以吗？\"\r\n\r\n### L2 进阶促成话术（利益汇总法）\r\n> \"李先生，我们回顾一下：您选择了[产品名称]，每年存[X]元，存[X]年，可以获得[核心保障]。今天投保的话，90天等待期后您的保障就生效了。现在是[当前月份]，正好是我们公司的[活动名称]，额外赠送[附加权益]。您看今天把合同签了？\"\r\n\r\n### L3 专家级促成话术（回马枪+压力促成）\r\n> \"李先生，我看您今天对这款产品确实有需求，也很认可我们的方案。\r\n> 我有个担心——您今天回去商量是应该的，但保险产品越早买越好，有两个原因：\r\n> 一是年龄越大保费越贵，每大一岁保费大约贵3%-5%；\r\n> 二是健康状况是不可逆的，如果体检查出什么异常，可能就买不了或需要加费。\r\n> 您今天要不要先把保费最低的方案确定下来？后续有任何调整我们再优化。\"\r\n\r\n### 评估标准\r\n- 🌟 优秀：能识别2个以上促成信号，并在最佳时机主动促成，话术自然不生硬\r\n- ✓ 合格：能识别一个促成信号，有基本促成尝试\r\n- ✗ 不合格：错过促成时机，或促成方式过于生硬让客户反感\r\n```\r\n\r\n---\r\n\r\n## 五、问题库统计表 / Question Bank Summary\r\n\r\n```markdown\r\n| 类别 | 基础⭐ | 入门⭐⭐ | 进阶⭐⭐⭐ | 高阶⭐⭐⭐⭐ | 专家⭐⭐⭐⭐⭐ | 小计 |\r\n|------|-------|--------|---------|-----------|-------------|------|\r\n| 产品知识（单选题） | 5 | 5 | 5 | 3 | 2 | 20 |\r\n| 产品知识（多选题） | 0 | 3 | 5 | 5 | 5 | 18 |\r\n| 计算题 | 0 | 0 | 3 | 5 | 5 | 13 |\r\n| 异议处理 | 5 | 5 | 5 | 3 | 2 | 20 |\r\n| 案例分析 | 0 | 0 | 5 | 5 | 5 | 15 |\r\n| 竞品对比 | 0 | 2 | 3 | 5 | 5 | 15 |\r\n| 促成话术 | 3 | 3 | 3 | 3 | 3 | 15 |\r\n| **合计** | **13** | **18** | **29** | **29** | **27** | **116** |\r\n```\r\n\r\n> 📌 每个产品建议生成 **100-150道** 题目，覆盖全部8个类别和5个难度级别\r\n> 📌 问题库建议每季度更新一次，结合实际销售数据和客户反馈\n\nFile v5.2.2:references/training_evaluation_rubric.md\n\n# Training Session Evaluation Rubric / 训练效果评估量表\r\n\r\n---\r\n\r\n## 一、综合评分体系 / Comprehensive Scoring System\r\n\r\n### 评分维度与权重\r\n\r\n| # | 评估维度 | 权重 | 说明 |\r\n|---|---------|------|------|\r\n| 1 | 产品知识准确率 | 25% | 产品条款、保障范围、数据记忆 |\r\n| 2 | 异议处理效果 | 25% | 反应速度、说服逻辑、共情能力 |\r\n| 3 | 需求挖掘深度 | 15% | 提问质量、需求识别精准度 |\r\n| 4 | 方案设计能力 | 15% | 产品组合逻辑、保费合理性 |\r\n| 5 | 合规话术规范 | 10% | 双录规范、禁止误导性表述 |\r\n| 6 | 沟通表达技巧 | 10% | 语言流畅度、客户参与感、节奏控制 |\r\n\r\n### 评分等级说明\r\n\r\n| 等级 | 分值区间 | 综合表现描述 |\r\n|------|---------|-------------|\r\n| 🌟 **A — 卓越** | 90-100分 | 全面超越L2标准，达到L3专家水平 |\r\n| ✅ **B — 优秀** | 80-89分 | 显著超越L2要求，能独立处理复杂场景 |\r\n| ⚡ **C — 良好** | 70-79分 | 达到L2标准，能独立完成标准销售流程 |\r\n| 📚 **D — 及格** | 60-69分 | L2基本要求，能在指导下完成销售 |\r\n| ❌ **F — 不及格** | <60分 | 未达到L2最低要求，需加强基础训练 |\r\n\r\n---\r\n\r\n## 二、分维度评估细则 / Dimension-Specific Rubrics\r\n\r\n### 2.1 产品知识准确率（25%）\r\n\r\n| 得分 | 表现描述 |\r\n|------|---------|\r\n| 23-25 | 全部产品条款准确无误，数据引用精确，能主动补充产品亮点 |\r\n| 18-22 | 主要条款准确（>90%），偶有小误差（缴费期/保障期），不影响说服力 |\r\n| 13-17 | 基本条款准确（>75%），有明显知识盲点（如特定条款/除外责任） |\r\n| 8-12 | 核心条款混淆，重要数据错误（2处以上） |\r\n| 0-7 | 完全无法准确描述产品，基础信息（名称/类型）都说不清 |\r\n\r\n### 2.2 异议处理效果（25%）\r\n\r\n**异议处理四维评估：**\r\n\r\n| 维度 | 权重 | 优秀（5分） | 良好（4分） | 及格（3分） | 不及格（1-2分） |\r\n|------|------|------------|------------|-----------|----------------|\r\n| **反应速度** | 25% | <15秒 | 15-30秒 | 30-60秒 | >60秒或无法回应 |\r\n| **说服逻辑** | 35% | 算账+共情+类比多管齐下 | 有逻辑，方法单一 | 有基本逻辑但说服力弱 | 逻辑混乱或直接否定客户 |\r\n| **共情表达** | 20% | 真诚共情，客户明显感到被理解 | 有共情但较形式化 | 偶尔共情，缺乏真诚 | 无共情，让客户感觉被反驳 |\r\n| **方案转化** | 20% | 将异议转化为成交机会 | 异议处理后推进方案 | 处理异议但未促成 | 处理完异议没有后续 |\r\n\r\n### 2.3 需求挖掘深度（15%）\r\n\r\n| 得分 | 表现描述 |\r\n|------|---------|\r\n| 14-15 | 能通过5+深度提问挖掘隐性需求（传承/税务/资产），并将其与产品关联 |\r\n| 11-13 | 能通过3-4个追问发现客户主要顾虑，设计针对性方案 |\r\n| 8-10 | 提问1-2个，以推销为主，客户参与度低 |\r\n| 5-7 | 基本不问需求，直接讲产品 |\r\n| 0-4 | 完全不问需求，单向灌输 |\r\n\r\n### 2.4 方案设计能力（15%）\r\n\r\n| 得分 | 表现描述 |\r\n|------|---------|\r\n| 14-15 | 3单以上产品组合，覆盖保障/储蓄/传承，逻辑清晰，保费占比合理（年收入10-20%） |\r\n| 11-13 | 2单产品组合，基本合理，保费有基本测算 |\r\n| 8-10 | 1-2单，保费测算不清晰或不合理（过高/过低） |\r\n| 5-7 | 仅有单品推荐，无组合意识 |\r\n| 0-4 | 无法设计任何可行方案 |\r\n\r\n---\r\n\r\n## 三、快问快答评估标准 / Quick Q&A Evaluation\r\n\r\n### 时间控制\r\n\r\n| 题数 | 建议时间 | 超时阈值 |\r\n|------|---------|---------|\r\n| 5题 | 5分钟 | >8分钟 |\r\n| 10题 | 10分钟 | >15分钟 |\r\n| 20题 | 20分钟 | >30分钟 |\r\n\r\n### 正确率评分\r\n\r\n| 正确率 | 评级 | 建议 |\r\n|--------|------|------|\r\n| ≥90% | 🌟 卓越 | 快速升级到高难度题 |\r\n| 75-89% | ✅ 优秀 | 保持当前难度，适度挑战 |\r\n| 60-74% | ⚡ 良好 | 巩固基础，减少失误 |\r\n| 40-59% | 📚 薄弱 | 回归基础题，加强产品手册学习 |\r\n| <40% | ❌ 不及格 | 暂停情景对练，专项补课产品知识 |\r\n\r\n---\r\n\r\n## 四、情景对练评估模板 / Role-Play Evaluation Template\r\n\r\n```markdown\r\n# 情景对练评估表\r\n\r\n代理人: ___________ 训练日期: ___________ 评估师: AI教练\r\n训练产品: ___________ 情景类型: ___________ 训练时长: _______分钟\r\n\r\n### 综合得分：_____ / 100\r\n\r\n| # | 评估项目 | 得分/10 | 优秀表现 | 待改进 |\r\n|---|---------|---------|---------|--------|\r\n| 1 | 开场破冰（建立信任） | ___/10 | | |\r\n| 2 | 需求挖掘提问质量 | ___/10 | | |\r\n| 3 | 产品讲解清晰度 | ___/10 | | |\r\n| 4 | 异议处理效果 | ___/20 | | |\r\n| 5 | 方案设计与推荐 | ___/15 | | |\r\n| 6 | 促成时机把握 | ___/10 | | |\r\n| 7 | 合规话术规范 | ___/10 | | |\r\n| 8 | 整体沟通流畅度 | ___/5 | | |\r\n| | **合计** | **___/100** | | |\r\n\r\n### 🌟 亮点时刻\r\n1. _________________________________\r\n\r\n### ⚠️ 改进重点\r\n1. _________________________________\r\n\r\n### 📋 下次训练建议\r\n- 产品：_____________\r\n- 场景：_____________\r\n- 时长：_____________\r\n```\r\n\r\n---\r\n\r\n## 五、训练效果追踪数据模型 / Training Progress Data Model\r\n\r\n```python\r\n# 训练效果追踪数据\r\n\r\ndef get_training_progress(agent_id: str) -> dict:\r\n return {\r\n \"agent_id\": agent_id,\r\n \"last_30_days\": {\r\n \"total_training_sessions\": 12,\r\n \"total_minutes\": 360,\r\n \"avg_score\": 74.2,\r\n \"score_trend\": \"+8.5\",\r\n \"sessions_by_type\": {\r\n \"quick_qa\": 5,\r\n \"role_play\": 4,\r\n \"case_study\": 2,\r\n \"assessment\": 1\r\n }\r\n },\r\n \"weak_point_progress\": {\r\n \"健康险异议处理\": {\"30d_ago\": 58, \"current\": 71, \"target\": 80, \"trend\": \"↑13\"},\r\n \"IRR计算\": {\"30d_ago\": 45, \"current\": 58, \"target\": 75, \"trend\": \"↑13\"}\r\n },\r\n \"recommended_next\": [\r\n {\r\n \"type\": \"情景对练\",\r\n \"product\": \"健康险\",\r\n \"focus\": \"'已有社保'异议场景专项训练\",\r\n \"reason\": \"本周得分71，距目标差9分，需强化\"\r\n }\r\n ]\r\n }\r\n```\n\nFile v5.2.2:skill-card.md\n\n## Description:
\nAI-powered insurance agent training coach that parses product documents, generates question banks, assesses agent skill levels, schedules personalized daily training, and supports interactive role-play sessions.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[gechengling](https://clawhub.ai/user/gechengling)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nInsurance trainers, agency leaders, and insurance technology teams use this skill to create coaching materials, question banks, competency assessments, daily training plans, and role-play practice for insurance agents. Outputs are advisory training materials that require qualified human review before use in sales, compliance, or product recommendations.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: The security evidence reports included Python scripts with external lottery API calls despite the skill text claiming there are no scripts or network calls.
\nMitigation: Review the scripts before installation, disable or remove lottery-data functions unless explicitly needed, and restrict outbound network access in regulated or enterprise deployments.
\nRisk: The skill can create insurance sales, compliance, and product recommendation training content that may be incorrect, outdated, or unsuitable for a specific jurisdiction or product.
\nMitigation: Require review by licensed insurance and compliance professionals before using generated materials with agents or customers.
\nRisk: Training scenarios may involve agent, client, schedule, or product information that could include personal or sensitive data.
\nMitigation: Do not provide real customer PII, use synthetic or redacted examples, and align any data handling with applicable privacy and insurance compliance requirements.
\n\n\n## Reference(s):
\n- [ClawHub skill page](https://clawhub.ai/gechengling/skills/insurance-agent-trainer)
\n- [Question bank templates](references/question_bank_templates.md)
\n- [Agent profile template](references/agent_profile_template.md)
\n- [Training evaluation rubric](references/training_evaluation_rubric.md)
\n\n\n## Skill Output:
\n**Output Type(s):** [Markdown, JSON, Analysis, Guidance, Configuration]
\n**Output Format:** [Markdown and JSON-style structured training artifacts]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [May include product profiles, question banks, competency assessments, daily training plans, role-play scripts, scoring rubrics, and human-review reminders.]
\n\n## Skill Version(s):
\n5.2.2 (source: frontmatter and server release evidence)
\n\n## Ethical Considerations:
\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.
\n\nArchive v5.2.1: 10 files, 41265 bytes\n\nFiles: README.md (6858b), references/agent_profile_template.md (9635b), references/question_bank_templates.md (13893b), references/training_evaluation_rubric.md (6398b), scripts/product_parser.py (8626b), scripts/question_generator.py (9212b), scripts/training_scheduler.py (11972b), skill-card.md (3047b), SKILL.md (26524b), _meta.json (142b)\n\nFile v5.2.1:SKILL.md\n\n---\nname: Insurance Agent Intelligent Trainer\ndescription: >\n AI-powered insurance agent training coach — auto-parses product docs, generates question banks,\n assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training\n based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers\n predinig rate cut (3.0%) impact on sales scripts, new health insurance regulation, PIPL-compliant\n customer communication, and benchmark against AIA, Ping An, and Alibaba Cloud training systems.\n Keywords: 智能陪练, 代理人培训, 保险训练, 产品陪练, 话术训练, 角色扮演, 技能评估, AIA培训, 平安培训, 代理人展业, 保险销售, 客户面谈, 异议处理, 保险培训系统, 学习路径.\nslug: insurance-agent-trainer\nversion: 5.2.1\nallowed-tools: []\ncapabilities:\n - educational-reference\n - advisory-only\n - requires-human-review\n - no-executable-code\n---\n\n# Insurance Agent Intelligent Trainer / 保险代理人智能陪练系统\n\n> **⚠️ SECURITY NOTICE / 安全声明**\n> - **Type:** Educational reference / analytical framework ONLY\n> - **No executable code, scripts, or binaries are included in this skill**\n> - **No persistent storage, network calls, background execution, or credential collection**\n> - **All outputs are for reference only and require human review before real-world application**\n> - **This skill does NOT provide financial, legal, or insurance advice**\n> - **Users must exercise their own judgment and consult qualified professionals**\n>\n> **⚠️ 数据安全警告**\n> - 本技能仅提供保险代理人的培训辅导参考框架，**不执行任何代码或脚本**\n> - 所有文档解析、日程分析、画像评估的描述均为**教学参考框架**，**不包含实际的OCR或PDF解析引擎**\n> - 不会自动访问、存储或处理用户的任何培训数据或个人信息\n> - 培训计划和话术建议需结合用户实际业务场景调整，**不能替代专业培训师**\n> - **销售话术和异议处理仅为培训参考，实际使用须遵守《保险法》及相关监管规定，不得以AI输出替代合规审核**\n\n### 保险监管最新动态 [2026-06-15更新]\n\n| 动态类型 | 内容摘要 | 发布时间 | 影响范围 |\n|---------|---------|---------|---------|\n| 监管发布 | NFRA 2026年第2号令：《银行保险机构许可证管理办法》6月1日施行 | 2026-06 | 代理人培训需新增许可证管理知识模块 |\n| 监管发布 | 许可证换证过渡期2026.6-2028.5，换证流程纳入培训 | 2026-06 | 代理人换证操作培训 |\n| 监管动态 | 2026年Q1监管处罚个人追责条款落地 | 2026-Q1 | 代理人合规培训需强化个人责任意识 |\n\n> **数据截止**: 2026-06-15 | 来源：国家金融监督管理总局、政府网、金融新闻网\n> **声明**: 以上动态供参考，具体以官方最新发布为准\n\n\n\n> **English:** AI-powered insurance agent coaching system — parses product documents, generates\n> personalized question banks, assesses agent competency levels, schedules daily training based on\n> client visits, and runs interactive role-play drills. Benchmarked against AIA, Ping An, and\n> Alibaba Cloud insurance training systems.\n>\n> **中文:** 保险代理人智能陪练系统——解析产品文档、自动生成问题库、评估代理人能力等级、\n> 结合当日客户拜访行程安排个性化训练、进行情景对练。对标友邦保险、平安保险、阿里云智能陪练水平。\n\n---\n\n## Trigger Keywords / 触发关键词\n\n**⚠️ 精确触发规则**：仅当用户明确提到保险代理人培训/陪练相关需求时激活。日常对话中提及\"培训\"、\"训练\"、\"coaching\"、\"agent training\"等通用词汇时**不会自动触发**。\n\n**用户确认规则**：当用户输入匹配以下关键词时，必须先确认用户意图：\n- \"您需要保险代理人陪练/培训服务吗？\"\n- 仅在用户明确确认后，才进入陪练模式\n\n激活关键词（需用户确认后生效）：\n\n- 保险陪练 / 产品陪练 / 智能陪练 / 代理人训练\n- 代理人培训 / 新人培训 / 保险话术训练\n- 产品演练 / 客户异议处理 / 保险销售训练\n- insurance agent training / insurance coaching / insurance product drill\n\n---\n\n## Core System Architecture / 核心系统架构\n\n### 0. 2025-2026 代理人销售环境最新变化\n\n| 变化 | 内容 | 话术调整建议 |\n|------|------|------------|\n| **预定利率降至3.0%** | 2024年9月后所有新产品执行 | 强调\"锁定3.0%长期确定收益\"，对比银行理财波动性 |\n| **分红险主导市场** | 分红险、万能险替代传统高利率产品 | 学会讲\"浮动收益+保底保障\"的双重价值 |\n| **健康险新规上线** | 2025年商业健康险管理办法修订 | 健康告知流程需更规范，禁止误导性说明 |\n| **代理人资格考试升级** | 2025年加入AI伦理、数字化服务模块 | 新人需补充数字化能力培训 |\n| **企微客户触达合规** | AI外呼需标注身份，营销需客户授权 | 培训合规营销话术，避免违规外呼 |\n\n\n\n```\n┌─────────────────────────────────────────────────────────────────┐\n│ Insurance Agent Intelligent Trainer │\n├─────────────────────────────────────────────────────────────────┤\n│ ┌──────────────┐ ┌──────────────┐ ┌──────────────────────┐ │\n│ │ Product Doc │ │ Agent Profile│ │ Daily Schedule/Routes│ │\n│ │ Parser │ │ Engine │ │ Integration │ │\n│ │ (PDF/Word/ │ │ (Skill Level │ │ (Today's Visits & │ │\n│ │ Images) │ │ Assessment) │ │ Client Profiles) │ │\n│ └──────┬───────┘ └──────┬───────┘ └──────────┬───────────┘ │\n│ │ │ │ │\n│ ▼ ▼ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Question Bank Generation Engine │ │\n│ │ Product Knowledge │ Objection Handling │ Case Analysis │ │\n│ │ [5 difficulty tiers × 3 categories = 15 question types] │ │\n│ └──────────────────────────┬───────────────────────────────┘ │\n│ │ │\n│ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Personalized Training Scheduler │ │\n│ │ [Skill Level + Schedule + Product Priority = Daily Plan]│ │\n│ └──────────────────────────┬───────────────────────────────┘ │\n│ │ │\n│ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Interactive Training Engine │ │\n│ │ Role-play │ Real-time Feedback │ Progress Tracking │ │\n│ └──────────────────────────────────────────────────────────┘ │\n└─────────────────────────────────────────────────────────────────┘\n```\n\n---\n\n## Core Capabilities / 核心能力\n\n### 1. Product Document Parser / 产品文档解析引擎（教学演示）\n\n> **⚠️ 教学演示**：以下展示产品文档解析的**概念性教学方法论**，仅说明AI可如何辅助理解产品结构。**本技能不执行任何实际的PDF解析、OCR识别或文档提取操作。** 所有\"解析流程\"均为逻辑示意，实际应用需由具体的工程实现完成。\n\n**Supported formats (conceptual):** PDF, Word (.docx), scanned images (with OCR), plain text\n\n**Conceptual parsing pipeline (for reference):**\n\n```\nDocument Upload\n │\n ▼\n[Format Detection] → PDF / Word / Image / Text\n │\n ▼\n[Text Extraction] → Raw text content\n │\n ▼\n[Structure Analysis]\n ├─ Product name, type, target customers\n ├─ Coverage scope (death, medical, annuity, critical illness, etc.)\n ├─ Premium levels & payment periods\n ├─ Policy terms & exclusions\n ├─ Sales pitch key points\n ├─ Competitive advantages vs. similar products\n └─ Compliance notes & regulatory requirements\n │\n ▼\n[Structured Product Profile] → Ready for question generation\n```\n\n**Output: Structured Product Profile JSON**\n\n```json\n{\n \"product_name\": \"XX福享人生终身寿险(万能型)\",\n \"product_type\": \"whole-life insurance with universal account\",\n \"insurer\": \"国联人寿\",\n \"target_customers\": [\"30-50岁中高收入人群\", \"有财富传承需求\"],\n \"coverage\": {\n \"death_benefit\": \"100%-160%账户价值\",\n \"annuity_option\": \"60岁起可转换为年金\",\n \"waiver\": \"可选投保人保费豁免\"\n },\n \"premium\": {\n \"min_annual\": 12000,\n \"payment_periods\": [\"3年\", \"5年\", \"10年\", \"20年\"],\n \"min_coverage_years\": \"终身\"\n },\n \"key_selling_points\": [\n \"复利增值，万能账户历史结算利率4.5%-5.2%\",\n \"灵活追加，额外资金可随时进入万能账户\",\n \"身故保障与财富传承双重功能\"\n ],\n \"competitive_edges\": [\"结算利率优于同类竞品\", \"追加无上限\"],\n \"exclusions\": [\"投保人对被保险人的故意伤害\", \"2年内自杀(无民事行为能力人除外)\"],\n \"compliance_notes\": [\"需双录(录音录像)\", \"犹豫期15天\", \"等待期90天\"],\n \"difficulty_tags\": [\"新人友好\", \"需强化健康告知\", \"财务规划综合能力\"]\n}\n```\n\n---\n\n### 2. Agent Profile & Skill Assessment / 代理人画像与能力评估\n\n> **⚠️ 数据处理提醒**：以下代理人画像和日程数据为**演示示例**。实际使用时，用户应自行管理代理人数据的收集和存储，确保符合《个人信息保护法》及保险行业合规要求。请勿输入真实客户PII信息。\n\n**Three skill tiers:**\n\n| Tier | Level | Description | Training Focus |\n|------|-------|-------------|----------------|\n| 🌱 **L1 - 入门级** | Beginner | < 1 year experience, struggles with product details and objection handling | Foundation: product knowledge, basic sales scripts, simple objection responses |\n| ⚡ **L2 - 进阶级** | Intermediate | 1-3 years, solid product knowledge but inconsistent closing rate | Application: complex scenarios, multi-product combination, competitive replacement, high-net-worth clients |\n| 🎯 **L3 - 专家级** | Advanced | 3+ years, high performance, needs strategy for complex cases | Mastery: enterprise/group clients, tax planning, estate planning, competitive stealing, mentoring skills |\n\n**Profile structure:**\n\n```json\n{\n \"agent_id\": \"AG20240001\",\n \"name\": \"张明\",\n \"level\": \"L2\",\n \"level_label\": \"进阶级\",\n \"tenure_years\": 2.5,\n \"certifications\": [\"保险代理人资格证\", \"健康险销售资质\"],\n \"performance\": {\n \"monthly_premium_target\": 50000,\n \"monthly_premium_actual\": 42000,\n \"closing_rate\": 0.32,\n \"avg_policy_size\": 18500,\n \"new_customer_rate\": 0.45\n },\n \"product_mastery\": {\n \"term_life\": 0.85,\n \"whole_life\": 0.72,\n \"critical_illness\": 0.58,\n \"medical_insurance\": 0.80,\n \"annuity\": 0.45,\n \"investment_linked\": 0.38\n },\n \"weak_points\": [\n \"健康险异议处理不够熟练\",\n \"不了解高端客户的税务筹划需求\",\n \"组合产品销售话术单一\"\n ],\n \"strong_points\": [\n \"老客户维护能力强\",\n \"缘故市场开拓优秀\"\n ],\n \"daily_schedule\": [\n {\"time\": \"09:00-10:00\", \"activity\": \"晨会\", \"location\": \"营业部\"},\n {\"time\": \"10:30-12:00\", \"activity\": \"拜访客户A（国企中层，有养老需求）\", \"location\": \"客户公司\"},\n {\"time\": \"14:00-15:30\", \"activity\": \"拜访客户B（私企业主，健康险需求）\", \"location\": \"客户公司\"},\n {\"time\": \"16:00-17:30\", \"activity\": \"缘故客户C（教育金规划）\", \"location\": \"咖啡厅\"}\n ]\n}\n```\n\n---\n\n### 3. Question Bank Generation / 问题库自动生成（教学模板）\n\n> **⚠️ 教学演示**：以下问题库和话术为**培训场景的教学参考模板**，展示如何结构化设计代理人训练内容。所有涉及销售话术、竞品对比、异议处理的内容均为**培训素材**，实际销售行为须遵循《保险法》及相关监管规定，并经持牌保险专业人士审核后方可执行。\n\n**Generated from product profile + agent level + training objectives**\n\n#### Question Types (15 categories across 3 dimensions)\n\n**By Category:**\n\n| Category | Description | Example |\n|----------|-------------|---------|\n| **产品知识** | Product features, terms, coverage | \"XX福的等待期是多久？\" |\n| **客户画像** | Target customer identification | \"什么样的客户适合购买这款产品？\" |\n| **异议处理** | Objection handling scripts | \"客户说'我已经有社保了，不需要商业保险'，如何回应？\" |\n| **案例分析** | Real case discussion | \"40岁国企中层，年薪50万，如何用这款产品做养老规划？\" |\n| **合规话术** | Compliance-approved scripts | \"如何向客户解释犹豫期和退保损失？\" |\n| **竞品对比** | vs. competitors | \"相比平安福，这款产品的核心优势是什么？\" |\n| **促成话术** | Closing techniques | \"客户表现出购买意向，如何自然促成？\" |\n| **交叉销售** | Multi-product combination | \"如何将主险与医疗险组合销售？\" |\n\n**By Difficulty (5 tiers):**\n\n| Level | Target Audience | Question Complexity |\n|--------|----------------|---------------------|\n| ⭐ 基础 | L1新人 | 单一产品，单一问题，直接答案 |\n| ⭐⭐ 入门 | L1-L2 | 单一产品，1-2个知识点，需要解释 |\n| ⭐⭐⭐ 进阶 | L2 | 单一产品，3-5个知识点，需组合分析 |\n| ⭐⭐⭐⭐ 高阶 | L2-L3 | 多产品组合，竞争替换，高净值客户 |\n| ⭐⭐⭐⭐⭐ 专家 | L3 | 综合方案，税务筹划，财富传承 |\n\n#### Question Bank Generation Prompt:\n\n```\nBased on the product profile provided, generate a question bank with:\n\n1. For each difficulty tier (基础/入门/进阶/高阶/专家):\n - 5 multiple choice questions (产品知识)\n - 3 case analysis questions\n - 3 objection handling scenarios\n - 2 competitive comparison questions\n - 1 closing technique exercise\n\n2. Total: 65+ questions per product\n\n3. For each question, provide:\n - Question text\n - Difficulty level (1-5)\n - Category (产品知识/异议处理/案例分析/竞品对比/促成话术)\n - Ideal answer / model response\n - Evaluation criteria (excellent/good/needs-improvement)\n - Coaching tips for the trainer\n```\n\n---\n\n### 4. Personalized Training Scheduler / 个性化训练调度引擎（方法论演示）\n\n> **⚠️ 教学演示**：以下调度算法、代理人画像及日程数据均为**教学方法论的概念性展示**。**本技能不实际采集、存储或处理任何代理人或客户数据**。所有姓名、日程、业绩数据均为虚构示例，仅用于说明逻辑框架。\n\n**Input factors:**\n\n```\nAgent Profile (Level + Weak Points)\n +\nToday's Client Schedule (Who → What need → What product)\n +\nProduct Priority Matrix\n =\nPersonalized Daily Training Plan\n```\n\n**Scheduling Algorithm:**\n\n```python\ndef generate_daily_training_plan(agent_profile, daily_schedule, products):\n \"\"\"\n Generate personalized training plan for the day.\n \"\"\"\n # Step 1: Identify today's client visit products\n today_products = extract_products_from_schedule(daily_schedule)\n \n # Step 2: Get agent's weakness areas for these products\n weakness_map = get_weakness_for_products(\n agent_profile.weak_points, \n today_products\n )\n \n # Step 3: Calculate training time available\n available_minutes = calculate_available_training_time(daily_schedule)\n \n # Step 4: Prioritize by impact × weakness × product value\n training_queue = prioritize_training(\n weakness_map,\n today_products,\n agent_profile.level,\n time_constraint=available_minutes\n )\n \n # Step 5: Generate session plan\n sessions = split_into_sessions(training_queue, available_minutes)\n \n return {\n \"date\": today,\n \"agent\": agent_profile.name,\n \"total_minutes\": available_minutes,\n \"sessions\": sessions,\n \"focus_products\": today_products,\n \"key_objectives\": get_key_objectives(training_queue)\n }\n```\n\n**Example Daily Training Plan:**\n\n```json\n{\n \"date\": \"2026-05-05\",\n \"agent\": \"张明\",\n \"level\": \"L2\",\n \"total_minutes\": 90,\n \"sessions\": [\n {\n \"time\": \"08:00-08:20\",\n \"duration\": 20,\n \"type\": \"晨间快练\",\n \"mode\": \"快问快答\",\n \"focus\": \"年金险产品知识（高频问题5题）\",\n \"product\": \"福享人生终身寿险\",\n \"objective\": \"巩固年金转换权的计算逻辑\"\n },\n {\n \"time\": \"12:30-13:00\",\n \"duration\": 30,\n \"type\": \"午间强化\",\n \"mode\": \"情景对练\",\n \"focus\": \"健康险异议处理\",\n \"scenario\": \"客户：\"我有社保，不需要商业医疗险\"\",\n \"product\": \"康健医疗保险\",\n \"level\": \"⭐⭐⭐ 进阶\",\n \"coaching_tips\": \"引导客户认识到社保报销比例上限，用自费药比例对比引发需求\"\n },\n {\n \"time\": \"17:30-18:30\",\n \"duration\": 40,\n \"type\": \"晚间复盘\",\n \"mode\": \"案例分析 + 角色扮演\",\n \"focus\": \"私企业主综合保障方案\",\n \"scenario\": \"45岁私企老板，年收入200万，已有多份保单，如何做加保方案？\",\n \"products\": [\"终身寿险+万能账户\", \"高端医疗\", \"企业财产险\"],\n \"level\": \"⭐⭐⭐⭐ 高阶\",\n \"model_response_guide\": \"从家庭资产与企业资产隔离角度切入，引出终身寿险的债务隔离和传承功能\"\n }\n ],\n \"key_metrics_to_track\": [\n \"异议处理响应时间（目标<30秒）\",\n \"产品知识点正确率（目标>85%）\",\n \"方案组合完整性（3单以上产品覆盖）\"\n ]\n}\n```\n\n---\n\n### 5. Interactive Training Session / 智能陪练对话引擎\n\n**Session modes:**\n\n| Mode | Description | Duration | Best For |\n|------|-------------|----------|----------|\n| **快问快答** | Rapid-fire Q&A | 5-10 min | Pre-meeting warmup |\n| **情景对练** | Role-play (client vs. agent) | 15-30 min | Skill practice |\n| **案例研讨** | Real case analysis | 20-40 min | Advanced agents |\n| **异议攻关** | Objection busting focus | 10-15 min | Weak point training |\n| **综合考核** | Full simulation exam | 30-60 min | Level assessment |\n\n**Real-time coaching during training:**\n\n```\nAgent Response\n │\n ▼\n[Natural Language Understanding] → Extract key claims, tone, strategy\n │\n ▼\n[Evaluation Engine]\n ├─ Product knowledge accuracy ✓/✗\n ├─ Objection handling effectiveness (1-5)\n ├─ Compliance adherence ✓/✗\n ├─ Closing attempt timing (good/early/late/missing)\n ├─ Client empathy signals ✓/✗\n └─ Product combination logic ✓/✗\n │\n ▼\n[Real-time Coaching Feedback]\n ├─ Immediate tip (if struggling): \"💡 提示：可以先问客户目前的保障缺口...\"\n ├─ Completion praise (if excellent): \"🌟 完美！您已经很好地识别了客户需求\"\n └─ Post-question summary: \"本轮得分 85/100。建议加强：竞品对比环节\"\n```\n\n**Training session flow:**\n\n```\n1. 导入 (5%) → 介绍训练目标和产品背景\n2. 暖场 (10%) → 快问快答热身，激活产品知识\n3. 主体 (60%) → 情景对练：客户角色扮演 + 实时点评\n4. 复盘 (20%) → AI给出详细反馈：优点/不足/改进建议\n5. 行动 (5%) → 下次拜访的具体行动计划\n```\n\n---\n\n### 6. Effect Assessment & Progress Tracking / 效果评估与进度追踪\n\n**Metrics tracked per session:**\n\n| Metric | Definition | Target |\n|--------|------------|--------|\n| **产品知识得分** | 知识点正确率 | L1: ≥70%, L2: ≥80%, L3: ≥90% |\n| **异议处理时效** | 从异议提出到满意回答的时间 | < 30秒 |\n| **促成成功率** | 能否自然引入促成信号 | ≥ 1次有效尝试 |\n| **话术合规率** | 合规敏感词使用正确性 | 100% |\n| **方案完整性** | 保障覆盖广度 | ≥ 3个维度 |\n\n**Progress report structure:**\n\n```markdown\n## 📊 代理人张明 训练报告 - 2026-05-05\n\n### 综合得分: ⭐⭐⭐⭐ (78/100)\n\n| 维度 | 本次得分 | 较上次 | 目标 |\n|------|---------|--------|------|\n| 产品知识 | 82/100 | ↑5 | 80+ |\n| 异议处理 | 71/100 | ↓3 | 75+ |\n| 促成技巧 | 85/100 | ↑8 | 80+ |\n| 合规话术 | 95/100 | →0 | 100 |\n| 方案设计 | 72/100 | ↑12 | 75+ |\n\n### 🔥 本次表现亮点\n1. 养老规划方案逻辑清晰，能结合客户生命周期讲解\n2. 合规话术使用规范，犹豫期/退保说明完整\n\n### ⚠️ 需要加强\n1. 健康险异议处理：回应\"已有社保\"时过于被动，应主动算账\n2. 竞品对比：对中国平安主要产品线不够熟悉\n\n### 📅 明日训练重点\n- 产品：康健医疗保险（健康告知流程）\n- 场景：竞品替换（平安福 vs. XX福）\n- 时长：30分钟情景对练 + 10分钟快问快答\n```\n\n---\n\n## Workflow / 标准工作流程\n\n> **⚠️ 重要提示**：以下工作流展示的是**培训场景的教学参考**。所有销售话术和异议处理内容均为培训素材，实际销售行为须遵循《保险法》及相关监管规定，经持牌保险专业人士审核。\n\n### Mode 1: Quick Start (已知产品 + 快速训练)\n\n```\nUser: \"帮我准备明天拜访客户B的训练，他是私企老板，对健康险感兴趣\"\n │\n ▼\n[Step 1] 获取代理人信息 → 张明，L2，弱项：健康险异议处理\n[Step 2] 识别拜访产品 → 康健医疗保险（目标：替换平安福）\n[Step 3] 生成训练计划 → 午间30分钟：健康险异议处理对练\n[Step 4] 开始陪练 → 情景对练：私企业主健康险需求挖掘\n[Step 5] 实时反馈 → 异议处理评分：71/100，给出改进建议\n[Step 6] 报告输出 → 训练报告 + 明日拜访话术优化建议\n```\n\n### Mode 2: Product Document Upload (上传产品文档)\n\n```\nUser: [上传 XX保险公司福享人生终身寿险 产品手册 PDF]\n │\n ▼\n[Step 1] 解析文档 → 提取产品结构、条款、卖点\n[Step 2] 生成产品画像 → Structured JSON Profile\n[Step 3] 生成问题库 → 65+道题目（5难度×8类别）\n[Step 4] 生成参考题库 → 作为AI对话上下文，**不持久存储**\n[Step 5] 等待选择 → \"请选择训练模式：快问快答 / 情景对练 / 案例研讨\"\n```\n\n### Mode 3: Full Agent Assessment (全面能力评估)\n\n```\nUser: \"帮我评估代理人李华的综合能力，她入职8个月，主要卖重疾险\"\n │\n ▼\n[Step 1] 建立代理人档案 → L1入门级，8个月，重疾险方向\n[Step 2] 产品文档上传 → 重疾险产品手册\n[Step 3] 综合考核 → 30题产品知识 + 5个情景对练\n[Step 4] 生成能力雷达图 → 6维度能力可视化\n[Step 5] 制定成长路径 → 90天训练计划\n```\n\n---\n\n## Input / Output Specifications / 输入输出规范\n\n### Input\n\n| Input Type | Description | Example |\n|------------|-------------|---------|\n| 代理人档案 | JSON/文本描述 | 姓名、级别、工龄、业绩、弱项 |\n| 产品文档 | PDF/Word/TXT/图片 | 保险产品手册、条款、计划书 |\n| 当日行程 | 文本/日历 | 09:00晨会 / 10:30拜访客户A |\n| 训练指令 | 自然语言 | \"帮我准备健康险的陪练\" |\n| 客户信息 | 文本描述 | \"45岁私企老板，年收入200万\" |\n\n### Output\n\n| Output Type | Description |\n|-------------|-------------|\n| 产品画像JSON | 结构化产品信息 |\n| 问题库 | 65+道分类分级题目 |\n| 训练计划 | 分钟级个性化日程 |\n| 陪练对话 | 实时AI角色扮演 |\n| 评估报告 | 评分 + 改进建议 + 雷达图 |\n| 成长路径 | 30/60/90天训练建议 |\n\n---\n\n## Integration Notes / 集成说明\n\n**Data privacy:**\n- All agent and client data remains local / within the company's system\n- No sensitive PII should be included in training documents\n- Comply with China CBIRC insurance sales compliance regulations\n\n**Lianxi with other Skills:**\n- `insurance-bidding-pro`: Use product analysis for bidding scenarios\n- `insurance-private-domain-ops`: Link training completion to customer follow-up\n- `insurance-claims-intelligence`: Train agents on claim processes for better client communication\n\n---\n\n## Disclaimer / 免责声明\n\n> ⚠️ **Training is advisory only.** This skill provides coaching materials, question banks,\n> and simulation training for insurance agent development. All final sales advice,\n> compliance decisions, and product recommendations must be reviewed by licensed\n> insurance professionals and comply with CBIRC regulations. Model answers represent\n> reference best practices, not guaranteed outcomes.\n\nFile v5.2.1:README.md\n\n# Insurance Agent Intelligent Trainer / 保险代理人智能陪练系统\r\n\r\n> **English** — AI-powered insurance agent coaching platform. Auto-parses product documents,\r\n> generates question banks, assesses agent competency (L1/L2/L3), schedules personalized daily\r\n> training based on client visits, and runs interactive role-play drills. Benchmarked against\r\n> AIA, Ping An, and Alibaba Cloud insurance training systems.\r\n\r\n> **中文** — 保险代理人智能陪练系统。自动解析产品文档、生成问题库、评估代理人能力等级、\r\n> 结合当日客户拜访行程安排个性化训练、进行情景对练。对标友邦保险、平安保险、阿里云智能陪练水平。\r\n\r\n---\r\n\r\n## ✨ Features / 核心功能\r\n\r\n### 🚀 Product Document Parser / 产品文档解析\r\n- **Input formats**: PDF, Word (.docx), scanned images (OCR), plain text\r\n- **Output**: Structured JSON product profile with coverage, terms, selling points, exclusions\r\n- **Official data**: Integrates with China Welfare Lottery & Sports Lottery public APIs for training case design\r\n\r\n### 🎯 Question Bank Generator / 问题库自动生成\r\n- **116+ questions** per product across **8 categories** × **5 difficulty tiers**\r\n- Categories: Product Knowledge · Objection Handling · Case Analysis · Competitive Comparison · Closing Techniques · Compliance Scripts · Needs Discovery · Digital Planning\r\n- Difficulty: ⭐基础 → ⭐⭐⭐⭐⭐专家 (L1–L3 agents)\r\n\r\n### 👤 Agent Profiling & Assessment / 代理人画像与评估\r\n- **3-tier competency model**: L1 (Beginner) / L2 (Intermediate) / L3 (Advanced)\r\n- **6-dimension radar chart**: Product Knowledge · Needs Discovery · Objection Handling · Closing · Compliance · Customer Relations\r\n- **Growth roadmap**: 30/60/90-day personalized development plans\r\n\r\n### 📅 Personalized Daily Training Scheduler / 个性化训练调度\r\n- Analyzes agent's daily client visit schedule\r\n- Maps visit products → training focus areas\r\n- Generates minute-level daily training plan with session recommendations\r\n\r\n### 🗣️ Interactive Training Modes / 智能陪练模式\r\n- **快问快答** — Rapid-fire Q&A warmup (5–10 min)\r\n- **情景对练** — Role-play (15–30 min)\r\n- **案例研讨** — Case analysis (20–40 min)\r\n- **异议攻关** — Objection busting focus (10–15 min)\r\n- **综合考核** — Full simulation exam (30–60 min)\r\n\r\n### 📊 Real-time Effect Tracking / 效果实时追踪\r\n- Session-level scoring (6 dimensions, 100-point scale)\r\n- 30-day trend analysis with improvement indicators\r\n- Radar chart visualization of competency progress\r\n\r\n---\r\n\r\n## 🚀 Quick Start\r\n\r\n### Installation / 安装\r\n\r\n```bash\r\n# Install via ClawHub\r\nopenclaw skills install insurance-agent-trainer\r\n\r\n# Or via npm\r\nnpx clawhub install @gechengling/insurance-agent-trainer\r\n```\r\n\r\n### Basic Usage / 基本使用\r\n\r\n#### 1. Upload product document and generate question bank\r\n```markdown\r\nUser: 请帮我解析[产品名称]的产品文档，并生成L2级别的问题库\r\n→ AI: 解析文档 → 生成116道题目 → 输出问题库JSON + Markdown报告\r\n```\r\n\r\n#### 2. Create agent profile and get daily training plan\r\n```markdown\r\nUser: 帮我安排代理人张明今天的训练计划，他今天要拜访3个客户（健康险+养老+教育金）\r\n→ AI: 分析行程 → 评估弱项 → 生成3段训练（共90分钟）\r\n```\r\n\r\n#### 3. Start interactive training session\r\n```markdown\r\nUser: 开始健康险的异议处理对练，我是L2级别\r\n→ AI: 启动情景对练 → AI扮演客户 → 实时点评 → 训练报告\r\n```\r\n\nArchive v5.2.0: 10 files, 41177 bytes\n\nFiles: README.md (6858b), references/agent_profile_template.md (9635b), references/question_bank_templates.md (13893b), references/training_evaluation_rubric.md (6398b), scripts/product_parser.py (8626b), scripts/question_generator.py (9212b), scripts/training_scheduler.py (11972b), skill-card.md (2915b), SKILL.md (26526b), _meta.json (142b)\n\nArchive v5.1.3: 10 files, 40873 bytes\n\nFiles: _meta.json (142b), README.md (6858b), references/agent_profile_template.md (9635b), references/question_bank_templates.md (13893b), references/training_evaluation_rubric.md (6398b), scripts/product_parser.py (8626b), scripts/question_generator.py (9212b), scripts/training_scheduler.py (11972b), skill-card.md (2964b), SKILL.md (25776b)\n\nArchive v5.1.2: 10 files, 40838 bytes\n\nFiles: README.md (6858b), references/agent_profile_template.md (9635b), references/question_bank_templates.md (13893b), references/training_evaluation_rubric.md (6398b), scripts/product_parser.py (8626b), scripts/question_generator.py (9212b), scripts/training_scheduler.py (11972b), skill-card.md (2868b), SKILL.md (25776b), _meta.json (142b)\n\nArchive v5.1.1: 10 files, 40623 bytes\n\nFiles: README.md (6858b), references/agent_profile_template.md (9635b), references/question_bank_templates.md (13893b), references/training_evaluation_rubric.md (6398b), scripts/product_parser.py (8626b), scripts/question_generator.py (9212b), scripts/training_scheduler.py (11972b), skill-card.md (2836b), SKILL.md (25186b), _meta.json (142b)\n\nArchive v5.1.0: 10 files, 40344 bytes\n\nFiles: README.md (6858b), references/agent_profile_template.md (9635b), references/question_bank_templates.md (13893b), references/training_evaluation_rubric.md (6398b), scripts/product_parser.py (8626b), scripts/question_generator.py (9212b), scripts/training_scheduler.py (11972b), skill-card.md (2984b), SKILL.md (24269b), _meta.json (142b)","readmeExcerpt":"Skill: Insurance Agent Intelligent Trainer Owner: gechengling Summary: AI-powered insurance agent training coach — auto-parses product docs, generates question banks, assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers predinig rate cut (3.0%) impact on sales scripts, new ","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"┌─────────────────────────────────────────────────────────────────┐\n│ Insurance Agent Intelligent Trainer │\n├─────────────────────────────────────────────────────────────────┤\n│ ┌──────────────┐ ┌──────────────┐ ┌──────────────────────┐ │\n│ │ Product Doc │ │ Agent Profile│ │ Daily Schedule/Routes│ │\n│ │ Parser │ │ Engine │ │ Integration │ │\n│ │ (PDF/Word/ │ │ (Skill Level │ │ (Today's Visits & │ │\n│ │ Images) │ │ Assessment) │ │ Client Profiles) │ │\n│ └──────┬───────┘ └──────┬───────┘ └──────────┬───────────┘ │\n│ │ │ │ │\n│ ▼ ▼ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Question Bank Generation Engine │ │\n│ │ Product Knowledge │ Objection Handling │ Case Analysis │ │\n│ │ [5 difficulty tiers × 3 categories = 15 question types] │ │\n│ └──────────────────────────┬───────────────────────────────┘ │\n│ │ │\n│ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Personalized Training Scheduler │ │\n│ │ [Skill Level + Schedule + Product Priority = Daily Plan]│ │\n│ └──────────────────────────┬───────────────────────────────┘ │\n│ │ │\n│ ▼ │\n│ ┌──────────────────────────────────────────────────────────┐ │\n│ │ Interactive Training Engine │ │\n│ │ Role-play │ Real-time Feedback │ Progress Tracking │ │\n│ └──────────────────────────────────────────────────────────┘ │\n└─────────────────────────────────────────────────────────────────┘"},{"language":"text","snippet":"Document Upload\n │\n ▼\n[Format Detection] → PDF / Word / Image / Text\n │\n ▼\n[Text Extraction] → Raw text content\n │\n ▼\n[Structure Analysis]\n ├─ Product name, type, target customers\n ├─ Coverage scope (death, medical, annuity, critical illness, etc.)\n ├─ Premium levels & payment periods\n ├─ Policy terms & exclusions\n ├─ Sales pitch key points\n ├─ Competitive advantages vs. similar products\n └─ Compliance notes & regulatory requirements\n │\n ▼\n[Structured Product Profile] → Ready for question generation"},{"language":"json","snippet":"{\n \"product_name\": \"XX福享人生终身寿险(万能型)\",\n \"product_type\": \"whole-life insurance with universal account\",\n \"insurer\": \"国联人寿\",\n \"target_customers\": [\"30-50岁中高收入人群\", \"有财富传承需求\"],\n \"coverage\": {\n \"death_benefit\": \"100%-160%账户价值\",\n \"annuity_option\": \"60岁起可转换为年金\",\n \"waiver\": \"可选投保人保费豁免\"\n },\n \"premium\": {\n \"min_annual\": 12000,\n \"payment_periods\": [\"3年\", \"5年\", \"10年\", \"20年\"],\n \"min_coverage_years\": \"终身\"\n },\n \"key_selling_points\": [\n \"复利增值，万能账户历史结算利率4.5%-5.2%\",\n \"灵活追加，额外资金可随时进入万能账户\",\n \"身故保障与财富传承双重功能\"\n ],\n \"competitive_edges\": [\"结算利率优于同类竞品\", \"追加无上限\"],\n \"exclusions\": [\"投保人对被保险人的故意伤害\", \"2年内自杀(无民事行为能力人除外)\"],\n \"compliance_notes\": [\"需双录(录音录像)\", \"犹豫期15天\", \"等待期90天\"],\n \"difficulty_tags\": [\"新人友好\", \"需强化健康告知\", \"财务规划综合能力\"]\n}"},{"language":"json","snippet":"{\n \"product_name\": \"XX康健终身重疾险(2026版)\",\n \"product_type\": \"critical illness insurance\",\n \"insurer\": \"国联人寿\",\n \"target_customers\": [\"28-50岁家庭经济支柱\", \"有重疾保障缺口人群\"],\n \"coverage\": {\n \"ci_types\": \"120种重疾+20种中症+40种轻症\",\n \"multiple_payout\": \"重疾1次+中症2次+轻症3次，累计最高260%保额\",\n \"death_benefit\": \"身故赔已交保费或现金价值较大者\"\n },\n \"premium\": {\n \"sample\": \"30岁男，50万保额，30年缴，年缴约 6800 元\",\n \"payment_periods\": [\"10年\",\"20年\",\"30年\"]\n },\n \"key_selling_points\": [\n \"重疾+中症+轻症三重递进保障\",\n \"轻中症豁免后续保费\",\n \"可附加恶性肿瘤二次赔付\"\n ],\n \"exclusions\": [\"投保前已患重疾\", \"遗传性疾病（条款约定）\", \"等待期内出险\"],\n \"compliance_notes\": [\"重疾定义以监管规范为准\", \"需明确告知等待期90-180天\", \"如实健康告知义务\"],\n \"difficulty_tags\": [\"健康告知敏感\", \"条款专业度高\", \"需结合医疗知识\"]\n}"},{"language":"json","snippet":"{\n \"agent_id\": \"AG20240001\",\n \"name\": \"张明\",\n \"level\": \"L2\",\n \"level_label\": \"进阶级\",\n \"tenure_years\": 2.5,\n \"certifications\": [\"保险代理人资格证\", \"健康险销售资质\"],\n \"performance\": {\n \"monthly_premium_target\": 50000,\n \"monthly_premium_actual\": 42000,\n \"closing_rate\": 0.32,\n \"avg_policy_size\": 18500,\n \"new_customer_rate\": 0.45\n },\n \"product_mastery\": {\n \"term_life\": 0.85,\n \"whole_life\": 0.72,\n \"critical_illness\": 0.58,\n \"medical_insurance\": 0.80,\n \"annuity\": 0.45,\n \"investment_linked\": 0.38\n },\n \"weak_points\": [\n \"健康险异议处理不够熟练\",\n \"不了解高端客户的税务筹划需求\",\n \"组合产品销售话术单一\"\n ],\n \"strong_points\": [\n \"老客户维护能力强\",\n \"缘故市场开拓优秀\"\n ],\n \"daily_schedule\": [\n {\"time\": \"09:00-10:00\", \"activity\": \"晨会\", \"location\": \"营业部\"},\n {\"time\": \"10:30-12:00\", \"activity\": \"拜访客户A（国企中层，有养老需求）\", \"location\": \"客户公司\"},\n {\"time\": \"14:00-15:30\", \"activity\": \"拜访客户B（私企业主，健康险需求）\", \"location\": \"客户公司\"},\n {\"time\": \"16:00-17:30\", \"activity\": \"缘故客户C（教育金规划）\", \"location\": \"咖啡厅\"}\n ]\n}"},{"language":"json","snippet":"{\n \"agent_id\": \"AG20230088\",\n \"name\": \"李华\",\n \"level\": \"L3\",\n \"level_label\": \"专家级\",\n \"tenure_years\": 6,\n \"certifications\": [\"保险代理人资格证\", \"CFP国际金融理财师\", \"私人银行家\"],\n \"performance\": {\n \"monthly_premium_target\": 200000,\n \"monthly_premium_actual\": 235000,\n \"closing_rate\": 0.48,\n \"avg_policy_size\": 86000,\n \"new_customer_rate\": 0.62\n },\n \"product_mastery\": {\n \"term_life\": 0.95, \"whole_life\": 0.92, \"critical_illness\": 0.90,\n \"medical_insurance\": 0.93, \"annuity\": 0.88, \"investment_linked\": 0.82\n },\n \"weak_points\": [\"家族信托等复杂传承架构经验不足\", \"跨境税务筹划需外部专家协同\"],\n \"strong_points\": [\"高净值客户经营\", \"企业团险开拓\", \"复杂方案设计\"],\n \"coaching_focus\": [\"传承架构进阶\", \"监管合规红线强化\", \"带教新人方法论\"]\n}"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: Insurance Agent Intelligent Trainer\ndescription: >\n AI-powered insurance agent training coach — auto-parses product docs, generates question banks,\n assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training\n based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers\n predinig rate cut (3.0%) impact on sales scripts, new health insurance regulation, PIPL-compliant\n customer communication, and benchmark against AIA, Ping An, and Alibaba Cloud training systems.\n Keywords: 智能陪练, 代理人培训, 保险训练, 产品陪练, 话术训练, 角色扮演, 技能评估, AIA培训, 平安培训, 代理人展业, 保险销售, 客户面谈, 异议处理, 保险培训系统, 学习路径.\nslug: insurance-agent-trainer\nversion: 5.2.5\nallowed-tools: []\ncapabilities:\n - educational-reference\n - advisory-only\n - requires-human-review\n - illustrative-code-samples\n---\n\n# Insurance Agent Intelligent Trainer / 保险代理人智能陪练系统\n\n> **⚠️ SECURITY NOTICE / 安全声明**\n> - **Type:** Educational reference / analytical framework ONLY\n> - **No runnable package is bundled:** the flow sketches, JSON samples and the short\n Python-style pseudocode below are illustrative reference material for you to adapt in\n your own environment; this skill ships no installer, service, parser or executable payload\n> - **No persistent storage, network calls, background execution, or credential collection**\n> - **All outputs are for reference only and require human review before real-world application**\n> - **This skill does NOT provide financial, legal, or insurance advice**\n> - **Users must exercise their own judgment and consult qualified professionals**\n>\n> **⚠️ 数据安全警告**\n> - 本技能仅提供保险代理人的培训辅导参考框架，**不内置可运行的解析器、题库引擎或存储服务**；\n 文中的流程示意、JSON 样例与伪代码均为教学参考，需由使用方在自己的系统中实现\n> - 所有文档解析、日程分析、画像评估的描述均为**教学参考框架**，**不包含实际的OCR或PDF解析引擎**\n> - 不会自动访问、存储或处理用户的任何培训数据或个人信息\n> - 培训计划和话术建议需结合用户实际业务场景调整，**不能替代专业培训师**\n> - **销售话术和异议处理仅为培训参考，实际使用须遵守《保险法》及相关监管规定，不得以AI输出替代合规审核**\n\n\n\n> **🔒 数据最小化前置声明（2026-09-24 新增，使用任何模块前先执行）**\n> 1. 只描述、不粘贴：客户信息以“45 岁私企老板、家庭年收入约 200 万”这类脱敏描述输入，禁止粘贴真实姓名、证件号、电话、住址、银行账号。\n> 2. 代理人档案同样脱敏：`agent_id` 用编号，姓名可用“张**”或化名。\n> 3. 训练记录默认只保留汇总指标（得分、弱项标签），不保留完整对话原文；确需保留须经本人同意并限定用途。\n> 4. 任何要写入文件或对外发送的训练报告、话术稿，先在对话中完整展示给用户预览，经明确确认后再落盘/发送。\n\n> **English:** AI-powered insurance agent coaching system — parses product documents, generates\n> personalized question banks, assesses agent competency levels, schedules daily training based on\n> client visits, and runs interactive role-play drills. Benchmarked against AIA, Ping An, and\n> Alibaba Cloud insurance training systems.\n>\n> **中文:** 保险代理人智能陪练系统——解析产品文档、自动生成问题库、评估代理人能力等级、\n> 结合当日客户拜访行程安排个性化训练、进行情景对练。对标友邦保险、平安保险、阿里云智能陪练水平。\n\n---\n\n## Trigger Keywords / 触发关键词\n\n**⚠️ 精确触发规则**：仅当用户明确提到保险代理人培训/陪练相关需求时激活。日常对话中提及\"培训\"、\"训练\"、\"coaching\"、\"agent training\"等通用词汇时**不会自动触发**。\n\n**用户确认规则**：当用户输入匹配以下关键词时，必须先确认用户意图：\n- \"您需要保险代理人陪练/培训服务吗？\"\n- 仅在用户明确确认后，才进入陪练模式\n\n激活关键词（需用户确认后生效）：\n\n- 保险陪练 / 产品陪练 / 智能陪练 / 代理人训练\n- 代理人培训 / 新人培训 / 保险话术训练\n- 产品演练 / 客户异议处理 / 保险销售训练\n- insurance agent training / insurance coaching / insurance p"},{"path":"_meta.json","content":"{\n \"ownerId\": \"kn74e704j3ygjcygnpf02rdvd185js13\",\n \"slug\": \"insurance-agent-trainer\",\n \"version\": \"5.2.5\",\n \"publishedAt\": 1790227032726\n}"},{"path":"skill-card.md","content":"## Description:\n\nAI-powered insurance agent training coach that parses product documents conceptually, generates question banks, assesses agent skill levels, schedules personalized daily training, and supports interactive role-play sessions.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[gechengling](https://clawhub.ai/user/gechengling)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nInsurance training teams, agency managers, and agents use this skill to prepare coaching materials, question banks, role-play drills, training plans, and advisory assessment reports for insurance-agent development. Outputs require human review before real sales, compliance, or product-recommendation use.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Training scripts or regulatory notes could be mistaken for current licensed insurance, legal, or financial advice.\n\nMitigation: Treat outputs as training examples only and require licensed compliance review before using generated material in real sales contexts.\n\nRisk: Agent or customer details entered during training could include sensitive personal information.\n\nMitigation: Avoid entering real customer PII, use desensitized descriptions, and keep any training records limited to summary metrics unless proper consent and controls are in place.\n\nRisk: Insurance rules, product terms, and sales-compliance requirements may change after the skill content was written.\n\nMitigation: Verify current insurance regulations and product terms against authoritative sources before applying any training output.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/gechengling/skills/insurance-agent-trainer)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, code, configuration, guidance]\n\n**Output Format:** [Markdown guidance with structured JSON examples and illustrative pseudocode.]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Advisory educational outputs; no executable package, persistent storage, network calls, or credential collection are bundled.]\n\n## Skill Version(s):\n\n5.2.5 (source: frontmatter and server release metadata)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment."}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"AI-powered insurance agent training coach — auto-parses product docs, generates question banks, assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers predinig rate cut (3.0%) impact on sales scripts, new health insurance regulation, PIPL-compliant customer communication, and benchmark against AIA, Ping An, and Alibaba Cloud training systems. Keywords: 智能陪练, 代理人培训, 保险训练, 产品陪练, 话术训练, 角色扮演, 技能评估, AIA培训, 平安培训, 代理人展业, 保险销售, 客户面谈, 异议处理, 保险培训系统, 学习路径. Skill: Insurance Agent Intelligent Trainer Owner: gechengling Summary: AI-powered insurance agent training coach — auto-parses product docs, generates question banks, assesses agent skill levels (beginner/intermediate/advanced), schedules personalized daily training based on client visits, and delivers interactive role-play sessions. Updated for 2025-2026: covers predinig rate cut (3.0%) impact on sales scripts, new","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1358,"uniquenessScore":49,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-09T23:27:51.973Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-09T23:27:51.973Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-10T03:02:14.696Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}