{
 "schema": "hit.v2",
 "generated_at": "2026-09-21T21:00:00+08:00",
 "method": "只读调用：POST http://127.0.0.1:8983/algo/<id>（本地算法调度器）。未改任何服务/cron/nginx/站点页面/数据库结构；未对外发布任何内容。每个配对都做了「换参数再调一次」的对照，参数不生效的单列。",
 "dispatcher": "http://127.0.0.1:8983",
 "public_entry": "https://shengminglian.com.cn/algo53/",
 "calls_total": 84,
 "pairs_total": 28,
 "algorithms_registered": 53,
 "algorithms_usable": 51,
 "algorithms_broken": [
  "spirit_engine",
  "value_discovery_engine"
 ],
 "algorithms_hit": [
  "business_seven_layers",
  "decay",
  "demand_verify",
  "five_dim_gold",
  "four_dim_landing",
  "funnel_evaluate",
  "industry_vitality",
  "k_factor",
  "motivation_level",
  "network_effect",
  "nine_axis_locator",
  "overlay_matching",
  "path_library",
  "pool_theory_eval",
  "pure_compound_six",
  "risk_grade",
  "six_line_five_color",
  "spread_validation",
  "three_d_profit",
  "traffic_target",
  "value_assessment_v123",
  "water_flow",
  "water_pool"
 ],
 "demands": {
  "D1": {
   "name": "北京·KTV 空档能不能开",
   "src": "geo/gap_top50.csv rank1 北京 KTV city_demand=0.962 方向=上行 cost=0.8 score=120.22",
   "note": "城市需求/方向/成本是真实台账数字；行业属性评分（自然属性/复利/刚需）无源数据，人工赋值，见\"要补什么\""
  },
  "D2": {
   "name": "上海·中医馆 是不是真需求",
   "src": "geo/gap_top50.csv rank15 上海 中医馆 city_demand=0.961 方向=上行 cost=0.8 score=120.10",
   "note": "同上；需求真伪的行为三项（搜过/试过/付过）无源数据，按\"未知→从最保守值\"赋值"
  },
  "D3": {
   "name": "深圳·咖啡店 落地可行性",
   "src": "geo/gap_top50.csv rank32 深圳 咖啡店 city_demand=0.918 方向=上行 cost=0.8 score=114.73",
   "note": "资金/时间/能力/风险四项无源数据，按行业常识中位赋值"
  },
  "D4": {
   "name": "重庆·健身房 风险等级",
   "src": "geo/gap_top50.csv rank44 重庆 健身房 city_demand=0.901 方向=上行 cost=0.8 score=112.66",
   "note": "无源 baseline；用 score 与其均值做基准偏离"
  },
  "D5": {
   "name": "民族服饰AI换装 出海首测$200",
   "src": "ads_scout §5 产品1 infogap.store/dress 首笔$200 看 CPA<=$4 止损 CPM>$40或CPA>$8",
   "note": "渠道单价来自报告实测自述（rd:p:krm9js CPM $15-25 / CPC<$1）"
  },
  "D6": {
   "name": "多语言方法论内容 newsletter $200",
   "src": "ads_scout §5 产品2 infogap.store/slices 首笔$200 投8-12个垂直newsletter 看注册成本<$2",
   "note": "newsletter CPM $250 来自 rd:p:1g7xuy3；$2000/20+刊 来自 rd:p:199rxrm"
  },
  "D7": {
   "name": "生命链平台 海外华人 Reddit冷启动$150",
   "src": "ads_scout §5 产品4 首笔$150 = $50 Reddit($5-10/天×7) + $100 两个newsletter 看留资成本<=$3",
   "note": "Reddit 实测：$700→77000曝光/180点击/0注册(rd:p:1vmp84p)；CPC$0.19 CPM$9.65 784点击1线索(rd:p:1253zfi)"
  },
  "D8": {
   "name": "TripCurve 行程规划 Google长尾$150",
   "src": "ads_scout §5 产品5 首笔$150($20/天×7 长尾词)+$200微网红 看每次生成<=$1.5 生成→注册>=8% 止损CPC>$3",
   "note": "Google CPC $0.4-2.6 来自报告 §2"
  },
  "D9": {
   "name": "分身引擎卖点 → 验证是否真需求（民族服饰AI换装）",
   "src": "commander_engine 分身引擎_算法版_卖点张老师 产出\"卖点\"；本需求=该卖点背后是不是真需求",
   "note": "卖点文案来自引擎产出场景，本机不打开引擎（只读场景定义）"
  },
  "D10": {
   "name": "分身引擎产出推广路径（重庆·健身房小老板）",
   "src": "commander_engine 分身引擎_舒老师_马师傅_卖点张老师 + 链指挥官v7 产出\"路径\"",
   "note": "人设九轴坐标无源数据，按台账 city_demand 与行业方向人工赋值"
  }
 },
 "hits": [
  {
   "demand_id": "D1",
   "demand_name": "北京·KTV 空档能不能开",
   "demand_src": "geo/gap_top50.csv rank1 北京 KTV city_demand=0.962 方向=上行 cost=0.8 score=120.22",
   "demand_note": "城市需求/方向/成本是真实台账数字；行业属性评分（自然属性/复利/刚需）无源数据，人工赋值，见\"要补什么\"",
   "algo": "five_dim_gold",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/five_dim_gold",
   "params": {
    "five_dim_market_size": 40,
    "five_dim_competition": 20,
    "five_dim_profit_model": 35,
    "five_dim_exec_difficulty": 30,
    "five_dim_sustainability": 25
   },
   "contrast_params": {
    "five_dim_market_size": 10,
    "five_dim_competition": 45,
    "five_dim_profit_model": 10,
    "five_dim_exec_difficulty": 5,
    "five_dim_sustainability": 45
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"module_id\": \"five_dim_gold\", \"name\": \"五维金衡商业评估\", \"dimensions\": {\"market_size\": {\"score\": 40, \"full\": 50, \"pct\": 80.0, \"label\": \"市场规模\"}, \"competition\": {\"score\": 20, \"full\": 50, \"pct\": 40.0, \"label\": \"竞争程度\"}, \"profit_model\": {\"score\": 35, \"full\": 50, \"pct\": 70.0, \"label\": \"盈利模式\"}, \"exec_difficulty\": {\"score\": 30, \"full\": 50, \"pct\": 60.0, \"label\": \"执行难度\"}, \"sustainability\": {\"score\": 25, \"full\": 50, \"pct\": 50.0, \"label\": \"可持续性\"}}, \"total_score\": 150, \"total_full\": 250, \"total_pct\": 60.0, \"long_term_value_raw\": -0.076, \"long_term_value_score\": 46.2, \"grade\": \"🟡B级\", \"grade_desc\": \"一般项目，需补齐短板\", \"suggestions\": [\"竞争程度偏低，需确认相对优势是否存在\", \"可持续性偏低，需构建护城河\", \"⚠️ 长期价值为负，执行难度与可持续性已侵蚀盈利，需重新评估\"], \"summary\": \"五维金衡总分150/250（60%）🟡B级，一般项目，需补齐短板\"}",
   "contrast_excerpt": "{\"module_id\": \"five_dim_gold\", \"name\": \"五维金衡商业评估\", \"dimensions\": {\"market_size\": {\"score\": 10, \"full\": 50, \"pct\": 20.0, \"label\": \"市场规模\"}, \"competition\": {\"score\": 45, \"full\": 50, \"pct\": 90.0, \"label\": \"竞争程度\"}, \"profit_model\": {\"score\": 10, \"full\": 50, \"pct\": 20.0, \"label\": \"盈利模式\"}, \"exec_difficulty\": {\"score\": 5, \"full\": 50, \"pct\": 10.0, \"label\": \"执行难度\"}, \"sustainability\": {\"score\": 45, \"full\": 50, \"pct\": 90.0, \"label\": \"可持续性\"}}, \"total_score\": 115, \"total_full\": 250, \"total_pct\": 46.0, \"long_term_value_raw\": -0.054, \"long_term_value_score\": 47.3, \"grade\": \"🔴C级\", \"grade_desc\": \"风险项目，建议放弃\", \"su",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "五维各 50 分 → 总 150/250、长期价值分 46.2、等级 B，并逐条给出短板建议。可整块塞进项目筛选页。"
  },
  {
   "demand_id": "D1",
   "demand_name": "北京·KTV 空档能不能开",
   "demand_src": "geo/gap_top50.csv rank1 北京 KTV city_demand=0.962 方向=上行 cost=0.8 score=120.22",
   "demand_note": "城市需求/方向/成本是真实台账数字；行业属性评分（自然属性/复利/刚需）无源数据，人工赋值，见\"要补什么\"",
   "algo": "pure_compound_six",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/pure_compound_six",
   "params": {
    "six_dim_frequency": 4,
    "six_dim_necessity": 2,
    "six_dim_margin": 4,
    "six_dim_lifetime": 3,
    "six_dim_standardization": 3,
    "six_dim_inheritability": 2
   },
   "contrast_params": {
    "six_dim_frequency": 1,
    "six_dim_necessity": 5,
    "six_dim_margin": 1,
    "six_dim_lifetime": 1,
    "six_dim_standardization": 5,
    "six_dim_inheritability": 5
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"module_id\": \"pure_compound_six\", \"name\": \"纯复利六维评估\", \"dimensions\": {\"frequency\": {\"score\": 4, \"full\": 5, \"label\": \"高频程度\", \"standard\": \"停用一个月难受=真高频\"}, \"necessity\": {\"score\": 2, \"full\": 5, \"label\": \"刚需程度\", \"standard\": \"提价50%还买=真刚需\"}, \"margin\": {\"score\": 4, \"full\": 5, \"label\": \"毛利空间\", \"standard\": \"CAC<30%首单=真高毛利\"}, \"lifetime\": {\"score\": 3, \"full\": 5, \"label\": \"终身属性\", \"standard\": \"主动续费率>90%=真终身\"}, \"standardization\": {\"score\": 3, \"full\": 5, \"label\": \"可标准化\", \"standard\": \"离开3个月还能运转=真标准化\"}, \"inheritability\": {\"score\": 2, \"full\": 5, \"label\": \"可传承性\", \"standard\": \"明天消失不影响=完全可传承\"}}, \"compound_index\": 576.0, \"max_possible\": 15625, \"compound_rate\": 3.69, \"avg_score\": 3.0, \"has_zero\": false, \"zero_conclusion\": \"\", \"grade\": \"🔴C级\", \"grade_desc\": \"线性模式，难以形成复利\", \"grade_emoji\": \"⚙️\", \"weaknesses\": [{\"dim\": \"necessity\", \"label\": \"刚需程度\", \"score\": 2, \"level\": \"需补强\"}, {\"dim\": \"inheritability\", \"label\": \"可传承性\", …[TRUNC]",
   "contrast_excerpt": "{\"module_id\": \"pure_compound_six\", \"name\": \"纯复利六维评估\", \"dimensions\": {\"frequency\": {\"score\": 1, \"full\": 5, \"label\": \"高频程度\", \"standard\": \"停用一个月难受=真高频\"}, \"necessity\": {\"score\": 5, \"full\": 5, \"label\": \"刚需程度\", \"standard\": \"提价50%还买=真刚需\"}, \"margin\": {\"score\": 1, \"full\": 5, \"label\": \"毛利空间\", \"standard\": \"CAC<30%首单=真高毛利\"}, \"lifetime\": {\"score\": 1, \"full\": 5, \"label\": \"终身属性\", \"standard\": \"主动续费率>90%=真终身\"}, \"standardization\": {\"score\": 5, \"full\": 5, \"label\": \"可标准化\", \"standard\": \"离开3个月还能运转=真标准化\"}, \"inheritability\": {\"score\": 5, \"full\": 5, \"label\": \"可传承性\", \"standard\": \"明天消失不影响=完全可传承\"}}, \"compound_index\": 125",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "六维各 5 分 → 复利指数 576/15625、has_zero 闸门、短板清单。与三维暴利互补（一个看赛道、一个看模型）。"
  },
  {
   "demand_id": "D1",
   "demand_name": "北京·KTV 空档能不能开",
   "demand_src": "geo/gap_top50.csv rank1 北京 KTV city_demand=0.962 方向=上行 cost=0.8 score=120.22",
   "demand_note": "城市需求/方向/成本是真实台账数字；行业属性评分（自然属性/复利/刚需）无源数据，人工赋值，见\"要补什么\"",
   "algo": "three_d_profit",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/three_d_profit",
   "params": {
    "three_d_profit_nature": 2,
    "three_d_profit_compound": 2,
    "three_d_profit_need": 3,
    "three_d_profit_details": "北京KTV: city_demand=0.962 上行 cost=0.8"
   },
   "contrast_params": {
    "three_d_profit_nature": 0,
    "three_d_profit_compound": 0,
    "three_d_profit_need": 0
   },
   "ok": true,
   "ms": 21,
   "output_excerpt": "{\"module_id\": \"three_d_profit\", \"name\": \"三维暴利指数\", \"nature_score\": 2, \"compound_score\": 2, \"need_score\": 3, \"profit_index\": 12, \"max_possible\": 27, \"profit_rate\": 44.44, \"abandon\": false, \"abandon_reason\": [], \"combo_type\": \"一高两中\", \"combo_desc\": \"有机会需运营\", \"combo_emoji\": \"📊\", \"typical\": \"保健品、订阅电商\", \"conclusion\": \"暴利指数 = 2×2×3 = 12（满分27）\\n暴利率 = 44.4%\\n组合类型：📊 一高两中 - 有机会需运营\\n典型赛道：保健品、订阅电商\", \"summary\": \"暴利指数=12/27，📊一高两中\"}",
   "contrast_excerpt": "{\"module_id\": \"three_d_profit\", \"name\": \"三维暴利指数\", \"nature_score\": 0, \"compound_score\": 0, \"need_score\": 0, \"profit_index\": 0, \"max_possible\": 27, \"profit_rate\": 0, \"abandon\": true, \"abandon_reason\": [\"自然属性为0，不顺应自然规律\", \"复利效应为0，无持续复购\", \"人人刚需为0，非必需\"], \"combo_type\": \"-\", \"combo_desc\": \"-\", \"combo_emoji\": \"❌\", \"typical\": \"-\", \"conclusion\": \"⚠️ 赛道放弃：至少一个维度为0，不具备暴利基础\", \"summary\": \"暴利指数=0/27，赛道放弃\"}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "输入三个 0-3 评分 → 输出暴利指数 12/27、暴利率 44.4%、组合类型「一高两中」、abandon 布尔闸门。相乘关系可解释，abandon=false 时可直接当「可以往下走」，true 时可直接砍。"
  },
  {
   "demand_id": "D10",
   "demand_name": "分身引擎产出推广路径（重庆·健身房小老板）",
   "demand_src": "commander_engine 分身引擎_舒老师_马师傅_卖点张老师 + 链指挥官v7 产出\"路径\"",
   "demand_note": "人设九轴坐标无源数据，按台账 city_demand 与行业方向人工赋值",
   "algo": "nine_axis_locator",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/nine_axis_locator",
   "params": {
    "axis_scores": {
     "capital": 2,
     "time": 3,
     "skill": 4,
     "stage": 3,
     "risk": 3,
     "era": 4,
     "region": 3,
     "network": 2,
     "health": 4
    },
    "unlocked_axes": 9
   },
   "contrast_params": {
    "axis_scores": {
     "capital": 5,
     "time": 1,
     "skill": 1,
     "stage": 5,
     "risk": 1,
     "era": 1,
     "region": 1,
     "network": 5,
     "health": 1
    },
    "unlocked_axes": 9
   },
   "ok": true,
   "ms": 664,
   "output_excerpt": "{\"module_id\": \"nine_axis_locator\", \"name\": \"九轴社会地图定位\", \"mode\": \"nine_axis_locator\", \"radar_data\": [{\"dimension\": \"资金\", \"key\": \"capital\", \"icon\": \"💰\", \"desc\": \"可支配资金与资产储备\", \"score\": 2, \"tag\": \"预算有限\", \"unlocked\": true}, {\"dimension\": \"时间\", \"key\": \"time\", \"icon\": \"⏰\", \"desc\": \"可自由支配的时间量\", \"score\": 3, \"tag\": \"时间一般\", \"unlocked\": true}, {\"dimension\": \"技能\", \"key\": \"skill\", \"icon\": \"🎯\", \"desc\": \"专业技能与核心竞争力\", \"score\": 4, \"tag\": \"技能扎实\", \"unlocked\": true}, {\"dimension\": \"阶段\", \"key\": \"stage\", \"icon\": \"📍\", \"desc\": \"人生阶段与当前处境\", \"score\": 3, \"tag\": \"发展期\", \"unlocked\": true}, {\"dimension\": \"风险\", \"key\": \"risk\", \"icon\": \"⚡\", \"desc\": \"风险承受与冒险倾向\", \"score\": 3, \"tag\": \"中性\", \"unlocked\": true}, {\"dimension\": \"年代\", \"key\": \"era\", \"icon\": \"📅\", \"desc\": \"年龄代际特征\", \"score\": 4, \"tag\": \"80后\", \"unlocked\": true}, {\"dimension\": \"体力\", \"key\": \"vitality\", \"icon\": \"💪\", \"desc\": \"身体精力与健康状态\", \"score\": 3, \"tag\": \"尚可\", \"unlocked\": tr …[TRUNC]",
   "contrast_excerpt": "{\"module_id\": \"nine_axis_locator\", \"name\": \"九轴社会地图定位\", \"mode\": \"nine_axis_locator\", \"radar_data\": [{\"dimension\": \"资金\", \"key\": \"capital\", \"icon\": \"💰\", \"desc\": \"可支配资金与资产储备\", \"score\": 5, \"tag\": \"资金充裕\", \"unlocked\": true}, {\"dimension\": \"时间\", \"key\": \"time\", \"icon\": \"⏰\", \"desc\": \"可自由支配的时间量\", \"score\": 1, \"tag\": \"时间匮乏\", \"unlocked\": true}, {\"dimension\": \"技能\", \"key\": \"skill\", \"icon\": \"🎯\", \"desc\": \"专业技能与核心竞争力\", \"score\": 1, \"tag\": \"技能待提升\", \"unlocked\": true}, {\"dimension\": \"阶段\", \"key\": \"stage\", \"icon\": \"📍\", \"desc\": \"人生阶段与当前处境\", \"score\": 5, \"tag\": \"收获期\", \"unlocked\": true}, {\"dimension\": \"风险\", \"key\": \"risk\",",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "九轴雷达 + 坐标 + 画像，耗时 664ms（本批唯一偏慢的）。可作免费入口的第一屏。"
  },
  {
   "demand_id": "D10",
   "demand_name": "分身引擎产出推广路径（重庆·健身房小老板）",
   "demand_src": "commander_engine 分身引擎_舒老师_马师傅_卖点张老师 + 链指挥官v7 产出\"路径\"",
   "demand_note": "人设九轴坐标无源数据，按台账 city_demand 与行业方向人工赋值",
   "algo": "overlay_matching",
   "round": "r3",
   "call": "POST http://127.0.0.1:8983/algo/overlay_matching",
   "params": {
    "coordinates": {
     "capital": 5,
     "time": 5,
     "skill": 5,
     "stage": 5,
     "risk": 5
    },
    "industry_tags": [
     "跨境",
     "SaaS"
    ],
    "skill": "编程",
    "funding": 5,
    "time_freedom": 5
   },
   "contrast_params": {
    "coordinates": {
     "capital": 1,
     "time": 1,
     "skill": 1,
     "stage": 1,
     "risk": 1
    },
    "industry_tags": [
     "健身"
    ],
    "skill": "健身",
    "funding": 1,
    "time_freedom": 1
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"quadrant\": \"综合型\", \"matches\": [{\"id\": \"G001\", \"name\": \"跨领域合作\", \"score\": 70.0}, {\"id\": \"G002\", \"name\": \"多元发展\", \"score\": 65.0}], \"count\": 2, \"top_opportunity\": {\"id\": \"G001\", \"name\": \"跨领域合作\", \"score\": 70.0}}",
   "contrast_excerpt": "{\"quadrant\": \"综合型\", \"matches\": [{\"id\": \"G001\", \"name\": \"跨领域合作\", \"score\": 67.3}, {\"id\": \"G002\", \"name\": \"多元发展\", \"score\": 62.3}], \"count\": 2, \"top_opportunity\": {\"id\": \"G001\", \"name\": \"跨领域合作\", \"score\": 67.3}}",
   "param_sensitive": true,
   "verdict": "废",
   "reason": "机会库只有 G001 跨领域合作 / G002 多元发展 两条，quadrant 恒为「综合型」；把坐标从 (5,5,5,5,5) 换成 (1,1,1,1,1) 也只把分数从 70.0 改到 67.3。这不是匹配，是两行常量。"
  },
  {
   "demand_id": "D10",
   "demand_name": "分身引擎产出推广路径（重庆·健身房小老板）",
   "demand_src": "commander_engine 分身引擎_舒老师_马师傅_卖点张老师 + 链指挥官v7 产出\"路径\"",
   "demand_note": "人设九轴坐标无源数据，按台账 city_demand 与行业方向人工赋值",
   "algo": "path_library",
   "round": "r2",
   "call": "POST http://127.0.0.1:8983/algo/path_library",
   "params": {
    "user_skills": [
     "短视频",
     "拍摄",
     "剪辑",
     "运营",
     "社群"
    ],
    "user_resources": [
     "本地人脉",
     "1家门店"
    ]
   },
   "contrast_params": {
    "user_skills": [
     "融资",
     "创业",
     "商业"
    ],
    "user_resources": []
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"matched_elves\": [{\"elf\": {\"id\": \"FY-VIDEO-001\", \"name\": \"短视频制作精灵\", \"skills\": [\"短视频\", \"拍摄\", \"剪辑\"]}, \"score\": 1.0}, {\"elf\": {\"id\": \"FY-COPY-001\", \"name\": \"文案精灵\", \"skills\": [\"文案\", \"创意\", \"内容\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-DESIGN-001\", \"name\": \"设计精灵\", \"skills\": [\"设计\", \"视觉\", \"品牌\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-LIVE-001\", \"name\": \"直播辅助机器人\", \"skills\": [\"直播\", \"带货\", \"互动\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-NETWORK-001\", \"name\": \"人脉精灵\", \"skills\": [\"人脉\", \"对接\", \"资源\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-AD-001\", \"name\": \"广告公司Agent\", \"skills\": [\"广告\", \"投放\", \"策略\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-COMMUNITY-001\", \"name\": \"社群运营精灵\", \"skills\": [\"社群\", \"运营\", \"促活\"]}, \"score\": 0.67}, {\"elf\": {\"id\": \"FY-INDUSTRY-001\", \"name\": \"行业精灵\", \"skills\": [\"行业\", \"趋势\", \"分析\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-DATA-001\", \"name\": \"记账数据精灵\", \"skills\": [\"记账\", \"数据\", \"分析\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-STARTUP-001\", \"name\": \"创业精灵\", \"skills\": [\"创业\", \"商业\", \"融资\"]}, \"score\": 0.3}], \"paths\": [{\"name\": \"短视频制作精灵专属路径\", \"elf\": {\"id\": \"FY-VIDEO-001\", \"name\": \"短视频制作精灵\", \"skills\": [\"短视频\", \"拍摄\", \"剪辑\"]}, \"match_score …[TRUNC]",
   "contrast_excerpt": "{\"matched_elves\": [{\"elf\": {\"id\": \"FY-VIDEO-001\", \"name\": \"短视频制作精灵\", \"skills\": [\"短视频\", \"拍摄\", \"剪辑\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-COPY-001\", \"name\": \"文案精灵\", \"skills\": [\"文案\", \"创意\", \"内容\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-DESIGN-001\", \"name\": \"设计精灵\", \"skills\": [\"设计\", \"视觉\", \"品牌\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-LIVE-001\", \"name\": \"直播辅助机器人\", \"skills\": [\"直播\", \"带货\", \"互动\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-NETWORK-001\", \"name\": \"人脉精灵\", \"skills\": [\"人脉\", \"对接\", \"资源\"]}, \"score\": 0.3}, {\"elf\": {\"id\": \"FY-AD-001\", \"name\": \"广告公司Agent\", \"skills\": [\"广告\", \"投放\", \"策略\"]}, \"score\": 0.3}, {\"elf\": {\"id\"",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "词表必须逐字命中精灵技能表：用 短视频/拍摄/剪辑/运营/社群 → 短视频精灵 1.0、社群精灵 0.67；换成 健身教练/门店运营 → 全部回落 0.3、输出看起来像废。对齐词表后可直接出路径。"
  },
  {
   "demand_id": "D2",
   "demand_name": "上海·中医馆 是不是真需求",
   "demand_src": "geo/gap_top50.csv rank15 上海 中医馆 city_demand=0.961 方向=上行 cost=0.8 score=120.10",
   "demand_note": "同上；需求真伪的行为三项（搜过/试过/付过）无源数据，按\"未知→从最保守值\"赋值",
   "algo": "demand_verify",
   "round": "r2",
   "call": "POST http://127.0.0.1:8983/algo/demand_verify",
   "params": {
    "answers": {
     "q1_self_statement": "上海中医馆有没有需求",
     "q2_searched": 1,
     "q3_tried": 0,
     "q4_paid": 0,
     "q5_willing": 2,
     "q6_action": 1
    }
   },
   "contrast_params": {
    "answers": {
     "q1_self_statement": "上海中医馆有没有需求",
     "q2_searched": 2,
     "q3_tried": 2,
     "q4_paid": 2,
     "q5_willing": 3,
     "q6_action": 3
    }
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"q1_self_statement\": \"上海中医馆有没有需求\", \"behavior_score\": 25, \"commitment_score\": 10.3, \"total_score\": 35.3, \"level\": \"🟠伪需求嫌疑\", \"confidence\": \"低\", \"behavior_details\": [\"✅ 主动搜索过解决方案(+25)\", \"❌ 未尝试过(0)\", \"❌ 未付过费(0)\", \"✅ 愿意采取具体行动(+10)\"], \"suggestion\": \"需进一步验证，问具体场景\"}",
   "contrast_excerpt": "{\"q1_self_statement\": \"上海中医馆有没有需求\", \"behavior_score\": 85, \"commitment_score\": 10.4, \"total_score\": 95.5, \"level\": \"🟢真需求\", \"confidence\": \"高\", \"behavior_details\": [\"✅ 主动搜索过解决方案(+25)\", \"✅ 尝试过其他方案(+30)\", \"✅ 为此付过费(+30)\", \"✅ 愿意采取具体行动(+10)\"], \"suggestion\": \"直接推进，提供解决方案\"}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "契约必须用 answers 嵌套 dict（顶层 q2_searched 等键会被忽略，返回恒 0 的「伪需求」）。改对后：行为 25 + 承诺 10.3 = 35.3 → 🟠伪需求嫌疑（低）；对照 95.5 → 🟢真需求（高）。是可直接当投放闸门用的分级。"
  },
  {
   "demand_id": "D2",
   "demand_name": "上海·中医馆 是不是真需求",
   "demand_src": "geo/gap_top50.csv rank15 上海 中医馆 city_demand=0.961 方向=上行 cost=0.8 score=120.10",
   "demand_note": "同上；需求真伪的行为三项（搜过/试过/付过）无源数据，按\"未知→从最保守值\"赋值",
   "algo": "industry_vitality",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/industry_vitality",
   "params": {
    "data_driven": 40,
    "digital_penetration": 35,
    "barrier": 70,
    "dim_scores": {
     "data_driven": 40,
     "substitutability": 30,
     "barrier": 70
    }
   },
   "contrast_params": {
    "data_driven": 95,
    "digital_penetration": 95,
    "barrier": 10,
    "dim_scores": {
     "data_driven": 95,
     "substitutability": 95,
     "barrier": 10
    }
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"dimensions\": [{\"id\": \"data_driven\", \"name\": \"数据驱动度\", \"desc\": \"行业决策是否依赖数据\", \"score\": 40, \"weight\": 0.4}, {\"id\": \"substitutability\", \"name\": \"可替代性\", \"desc\": \"被技术/新模式替代的风险\", \"score\": 30, \"weight\": 0.3}, {\"id\": \"barrier\", \"name\": \"进入壁垒\", \"desc\": \"行业护城河深度\", \"score\": 70, \"weight\": 0.3}], \"weighted_score\": 46.0, \"classification\": \"🟡数据可替代型\", \"description\": \"数字化过渡期行业，面临被替代风险需转型\", \"vitality_level\": \"中\"}",
   "contrast_excerpt": "{\"dimensions\": [{\"id\": \"data_driven\", \"name\": \"数据驱动度\", \"desc\": \"行业决策是否依赖数据\", \"score\": 95, \"weight\": 0.4}, {\"id\": \"substitutability\", \"name\": \"可替代性\", \"desc\": \"被技术/新模式替代的风险\", \"score\": 95, \"weight\": 0.3}, {\"id\": \"barrier\", \"name\": \"进入壁垒\", \"desc\": \"行业护城河深度\", \"score\": 10, \"weight\": 0.3}], \"weighted_score\": 69.5, \"classification\": \"🟡数据可替代型\", \"description\": \"数字化过渡期行业，面临被替代风险需转型\", \"vitality_level\": \"中\"}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "三维加权 → 46 分、🟡数据可替代型。壁垒给到 70 仍被拉回「可替代」，说明它真的在算权重而不是念稿。可作行业页结论。"
  },
  {
   "demand_id": "D2",
   "demand_name": "上海·中医馆 是不是真需求",
   "demand_src": "geo/gap_top50.csv rank15 上海 中医馆 city_demand=0.961 方向=上行 cost=0.8 score=120.10",
   "demand_note": "同上；需求真伪的行为三项（搜过/试过/付过）无源数据，按\"未知→从最保守值\"赋值",
   "algo": "value_assessment_v123",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/value_assessment_v123",
   "params": {
    "product_name": "上海中医馆",
    "product_desc": "上海中医馆门店服务；geo缺口rank15 city_demand=0.961 上行",
    "efficiency_gain": 6,
    "cost_reduction": 5,
    "coverage_scale": 5,
    "substitution_risk": 0.4,
    "security_score": 8,
    "pleasure_score": 4,
    "showcase_score": 3,
    "emotion_duration": 0.6,
    "secondary_market": 2,
    "consensus_strength": 7,
    "value_period": 0.7,
    "scarcity": 6
   },
   "contrast_params": {
    "product_name": "上海中医馆",
    "product_desc": "x",
    "efficiency_gain": 1,
    "cost_reduction": 1,
    "coverage_scale": 1,
    "substitution_risk": 0.9,
    "security_score": 1,
    "pleasure_score": 1,
    "showcase_score": 1,
    "emotion_duration": 0.1,
    "secondary_market": 1,
    "consensus_strength": 1,
    "value_period": 0.1,
    "scarcity": 1
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"module\": \"真需求价值评估 V1~V3\", \"product_name\": \"上海中医馆\", \"product_desc\": \"上海中医馆门店服务；geo缺口rank15 city_demand=0.961 上行\", \"v1_functional\": {\"score\": 0.94, \"base_score\": 1.0, \"model_bonus\": 0, \"risk_penalty\": 0.06, \"model_identified\": {\"model\": \"infrastructure\", \"name\": \"基础设施模型\", \"confidence\": \"高\"}, \"model_input\": \"\", \"substitution_risk\": 0.4, \"level\": \"🟢极高\", \"eval\": \"功能价值极高\"}, \"v2_emotional\": {\"score\": 0.88, \"base_score\": 1.0, \"duration_factor\": 0.88, \"security\": 1, \"pleasure\": 1, \"showcase\": 1, \"dominant_element\": {\"id\": \"security\", \"name\": \"保障感\", \"score\": 1, \"desc\": \"对抗害怕\"}, \"level\": \"🟢极高\", \"eval\": \"情绪价值极强\"}, \"v3_asset\": {\"score\": 0.8, \"base_score\": 1.0, \"period_factor\": 0.5, \"secondary_market\": 1, \"consensus_strength\": 1, \"scarcity\": 1, \"value_period_years\": 0.7, \"has_market\": true, \"has_consensus\": true, \"level\": \"🟢高\", \"eval\": \"资产价值良好\"}, \"F_score\": 0.94, \"E_score\": 0.88, \"A_score\": 0.8, \"V_ …[TRUNC]",
   "contrast_excerpt": "{\"module\": \"真需求价值评估 V1~V3\", \"product_name\": \"上海中医馆\", \"product_desc\": \"x\", \"v1_functional\": {\"score\": 0.865, \"base_score\": 1.0, \"model_bonus\": 0, \"risk_penalty\": 0.135, \"model_identified\": {\"model\": \"infrastructure\", \"name\": \"基础设施模型\", \"confidence\": \"高\"}, \"model_input\": \"\", \"substitution_risk\": 0.9, \"level\": \"🟢极高\", \"eval\": \"功能价值极高\"}, \"v2_emotional\": {\"score\": 0.73, \"base_score\": 1.0, \"duration_factor\": 0.73, \"security\": 1, \"pleasure\": 1, \"showcase\": 1, \"dominant_element\": {\"id\": \"security\", \"name\": \"保障感\", \"score\": 1, \"desc\": \"对抗害怕\"}, \"level\": \"🟢高\", \"eval\": \"情绪价值良好\"}, \"v3_asset\": {\"score\": 0.8, \"",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "V1 0.94 / V2 0.88 / V3 0.80 + 自动识别「基础设施模型」+ V_total + value_type，直接回答「这东西值不值得做」。输入稍有不同分数就变，不是摆设。"
  },
  {
   "demand_id": "D3",
   "demand_name": "深圳·咖啡店 落地可行性",
   "demand_src": "geo/gap_top50.csv rank32 深圳 咖啡店 city_demand=0.918 方向=上行 cost=0.8 score=114.73",
   "demand_note": "资金/时间/能力/风险四项无源数据，按行业常识中位赋值",
   "algo": "four_dim_landing",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/four_dim_landing",
   "params": {
    "four_dim_fund": 15,
    "four_dim_time": 18,
    "four_dim_ability": 14,
    "four_dim_risk": 12
   },
   "contrast_params": {
    "four_dim_fund": 25,
    "four_dim_time": 5,
    "four_dim_ability": 25,
    "four_dim_risk": 25
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"module_id\": \"four_dim_landing\", \"name\": \"四维度落地评估\", \"dimensions\": {\"fund\": {\"score\": 15, \"full\": 25, \"pct\": 60.0, \"label\": \"资金\"}, \"time\": {\"score\": 18, \"full\": 25, \"pct\": 72.0, \"label\": \"时间\"}, \"ability\": {\"score\": 14, \"full\": 25, \"pct\": 56.0, \"label\": \"能力\"}, \"risk\": {\"score\": 12, \"full\": 25, \"pct\": 48.0, \"label\": \"风险（反向）\"}}, \"risk_positive\": 13, \"total_score\": 60, \"total_full\": 100, \"landing_index\": 60.0, \"grade\": \"🟡B级\", \"grade_desc\": \"落地可行性一般，需补齐短板\", \"grade_emoji\": \"🟡\", \"weaknesses\": [{\"dim\": \"fund\", \"label\": \"资金\", \"score\": 15, \"full\": 25, \"level\": \"需补强\", \"question\": \"是否有足够资金启动和维持？\"}, {\"dim\": \"ability\", \"label\": \"能力\", \"score\": 14, \"full\": 25, \"level\": \"需补强\", \"question\": \"是否具备所需能力？\"}, {\"dim\": \"risk\", \"label\": \"风险\", \"score\": 12, \"full\": 25, \"level\": \"需补强\", \"question\": \"风险是否可控？\"}], \"suggestions\": [\"能力不足，建议补充所需技能或寻找合作伙伴\", \"💡 补齐短板后可启动\"], \"summary\": \"落地评估60/100🟡 🟡B级 | 资金15 时间18 能力14 风险12(反向) …[TRUNC]",
   "contrast_excerpt": "{\"module_id\": \"four_dim_landing\", \"name\": \"四维度落地评估\", \"dimensions\": {\"fund\": {\"score\": 25, \"full\": 25, \"pct\": 100.0, \"label\": \"资金\"}, \"time\": {\"score\": 5, \"full\": 25, \"pct\": 20.0, \"label\": \"时间\"}, \"ability\": {\"score\": 25, \"full\": 25, \"pct\": 100.0, \"label\": \"能力\"}, \"risk\": {\"score\": 25, \"full\": 25, \"pct\": 100.0, \"label\": \"风险（反向）\"}}, \"risk_positive\": 0, \"total_score\": 55, \"total_full\": 100, \"landing_index\": 55.0, \"grade\": \"🟡B级\", \"grade_desc\": \"落地可行性一般，需补齐短板\", \"grade_emoji\": \"🟡\", \"weaknesses\": [{\"dim\": \"time\", \"label\": \"时间\", \"score\": 5, \"full\": 25, \"level\": \"严重短板\", \"question\": \"是否有足够时间投入？\"}, {\"dim\": ",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "资金/时间/能力/风险各 25 → 落地指数 60、B 级，并输出逐维「需补强」+ 追问句。可直接进落地评估页。"
  },
  {
   "demand_id": "D3",
   "demand_name": "深圳·咖啡店 落地可行性",
   "demand_src": "geo/gap_top50.csv rank32 深圳 咖啡店 city_demand=0.918 方向=上行 cost=0.8 score=114.73",
   "demand_note": "资金/时间/能力/风险四项无源数据，按行业常识中位赋值",
   "algo": "pool_theory_eval",
   "round": "r3",
   "call": "POST http://127.0.0.1:8983/algo/pool_theory_eval",
   "params": {
    "income_total": 200000,
    "expense_total": 50000,
    "asset_total": 3000000
   },
   "contrast_params": {
    "income_total": 1000,
    "expense_total": 90000,
    "asset_total": 1000
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"mode\": \"data\", \"module_id\": \"pool_theory_eval\", \"name\": \"水池理论诊断评估\", \"total_income\": 200000, \"total_expense\": 50000, \"net_flow\": 150000, \"saving_rate\": 75.0, \"pool_capacity\": 0.0, \"pool_fullness\": 0.0, \"coverage_months\": 0.0, \"compound_rate\": 0.0, \"compound_effect\": 0.0, \"asset_growth\": 150000.0, \"inlet_quality\": \"充足\", \"outlet_quality\": \"健康\", \"quadrant\": \"有隐患\", \"quadrant_emoji\": \"🟡\", \"strategy\": \"查出水口堵漏\", \"risk_level\": \"中低风险\", \"trend\": \"增长↑\", \"trend_emoji\": \"📈\", \"inlet_count\": 0, \"outlet_count\": 0, \"summary\": \"🟡 有隐患 | 收入200000-支出50000=净流150000 | 结余率75% | 📈增长↑ | 查出水口堵漏\"}",
   "contrast_excerpt": "{\"mode\": \"data\", \"module_id\": \"pool_theory_eval\", \"name\": \"水池理论诊断评估\", \"total_income\": 1000, \"total_expense\": 90000, \"net_flow\": -89000, \"saving_rate\": 0, \"pool_capacity\": 0.0, \"pool_fullness\": 0.0, \"coverage_months\": 0.0, \"compound_rate\": 0.0, \"compound_effect\": 0.0, \"asset_growth\": -89000.0, \"inlet_quality\": \"不足\", \"outlet_quality\": \"漏血\", \"quadrant\": \"危急\", \"quadrant_emoji\": \"🔴\", \"strategy\": \"紧急止血+重建\", \"risk_level\": \"极高风险\", \"trend\": \"缩减↓\", \"trend_emoji\": \"📉\", \"inlet_count\": 0, \"outlet_count\": 0, \"summary\": \"🔴 危急 | 收入1000-支出90000=净流-89000 | 结余率0% | 📉缩减↓ | 紧急止血+重建\"}",
   "param_sensitive": true,
   "verdict": "勉强",
   "reason": "数据模式算得干净（净流 150000、结余率 75%、象限「有隐患」🟡）。但阈值偏激进：结余率 10% 就判「危急/极高风险」，而满配也只到「有隐患」——没有 🟢 档。结论色不能直接给用户看。"
  },
  {
   "demand_id": "D3",
   "demand_name": "深圳·咖啡店 落地可行性",
   "demand_src": "geo/gap_top50.csv rank32 深圳 咖啡店 city_demand=0.918 方向=上行 cost=0.8 score=114.73",
   "demand_note": "资金/时间/能力/风险四项无源数据，按行业常识中位赋值",
   "algo": "six_line_five_color",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/six_line_five_color",
   "params": {
    "user_data": {
     "age": 32,
     "profession": "餐饮店长",
     "income": 9000,
     "region_score": 70
    },
    "age": 32,
    "profession": "餐饮店长",
    "income": 9000,
    "region_score": 70
   },
   "contrast_params": {
    "user_data": {
     "age": 55,
     "profession": "退休",
     "income": 2000,
     "region_score": 20
    },
    "age": 55,
    "profession": "退休",
    "income": 2000,
    "region_score": 20
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"scores\": {\"时间自由度\": 60, \"资金\": 30, \"技能\": 70, \"地域\": 70, \"生命阶段\": 70, \"体力\": 70}, \"color\": \"黄色\", \"tags\": [\"餐饮店长\", \"黄色阶段\"], \"total\": 370}",
   "contrast_excerpt": "{\"scores\": {\"时间自由度\": 60, \"资金\": 30, \"技能\": 70, \"地域\": 20, \"生命阶段\": 60, \"体力\": 70}, \"color\": \"黄色\", \"tags\": [\"退休\", \"黄色阶段\"], \"total\": 310}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "六维分随 user_data 真变（32岁/餐饮店长 → 时间自由度 60、资金 30、技能 70；55岁/退休 → 另一组），输出颜色 + 标签。可作画像卡。"
  },
  {
   "demand_id": "D4",
   "demand_name": "重庆·健身房 风险等级",
   "demand_src": "geo/gap_top50.csv rank44 重庆 健身房 city_demand=0.901 方向=上行 cost=0.8 score=112.66",
   "demand_note": "无源 baseline；用 score 与其均值做基准偏离",
   "algo": "business_seven_layers",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/business_seven_layers",
   "params": {
    "seven_layer_math_thinking": 70,
    "seven_layer_values": 75,
    "seven_layer_project_screening": 80,
    "seven_layer_math_decision": 60,
    "seven_layer_landing_feasibility": 55,
    "seven_layer_strategy": 65,
    "seven_layer_profit_system": 60
   },
   "contrast_params": {
    "seven_layer_math_thinking": 20,
    "seven_layer_values": 20,
    "seven_layer_project_screening": 20,
    "seven_layer_math_decision": 20,
    "seven_layer_landing_feasibility": 20,
    "seven_layer_strategy": 20,
    "seven_layer_profit_system": 20
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"module_id\": \"business_seven_layers\", \"name\": \"商业7层体系\", \"layers\": [{\"key\": \"math_thinking\", \"name\": \"数学思维\", \"description\": \"用数学逻辑理解商业本质的能力\", \"score\": 70, \"full\": 100, \"pct\": 70, \"grade\": \"🟡B级\", \"grade_emoji\": \"🟡\", \"order\": 1}, {\"key\": \"values\", \"name\": \"价值观\", \"description\": \"商业伦理与价值判断\", \"score\": 75, \"full\": 100, \"pct\": 75, \"grade\": \"⭐A级\", \"grade_emoji\": \"⭐\", \"order\": 2}, {\"key\": \"project_screening\", \"name\": \"项目筛选\", \"description\": \"基于标准筛选和评估项目的能力\", \"score\": 80, \"full\": 100, \"pct\": 80, \"grade\": \"⭐A级\", \"grade_emoji\": \"⭐\", \"order\": 3}, {\"key\": \"math_decision\", \"name\": \"数学决策\", \"description\": \"基于数据的理性决策能力\", \"score\": 60, \"full\": 100, \"pct\": 60, \"grade\": \"🟡B级\", \"grade_emoji\": \"🟡\", \"order\": 4}, {\"key\": \"landing_feasibility\", \"name\": \"落地可行性\", \"description\": \"将决策转化为行动的能力\", \"score\": 55, \"full\": 100, \"pct\": 55, \"grade\": \"🔴C级\", \"grade_emoji\": \"🔴\", \"order\": 5}, {\"key\": \"strategy\", \"name\": \"战略制定\", \"des …[TRUNC]",
   "contrast_excerpt": "{\"module_id\": \"business_seven_layers\", \"name\": \"商业7层体系\", \"layers\": [{\"key\": \"math_thinking\", \"name\": \"数学思维\", \"description\": \"用数学逻辑理解商业本质的能力\", \"score\": 20, \"full\": 100, \"pct\": 20, \"grade\": \"🔴C级\", \"grade_emoji\": \"🔴\", \"order\": 1}, {\"key\": \"values\", \"name\": \"价值观\", \"description\": \"商业伦理与价值判断\", \"score\": 20, \"full\": 100, \"pct\": 20, \"grade\": \"🔴C级\", \"grade_emoji\": \"🔴\", \"order\": 2}, {\"key\": \"project_screening\", \"name\": \"项目筛选\", \"description\": \"基于标准筛选和评估项目的能力\", \"score\": 20, \"full\": 100, \"pct\": 20, \"grade\": \"🔴C级\", \"grade_emoji\": \"🔴\", \"order\": 3}, {\"key\": \"math_decision\", \"name\": \"数学决策\", \"description\": \"基于数据",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "七层各 100 分 → 逐层 A/B 级 + 总分 + 短板 + 建议。适合当「老板成熟度」总表。"
  },
  {
   "demand_id": "D4",
   "demand_name": "重庆·健身房 风险等级",
   "demand_src": "geo/gap_top50.csv rank44 重庆 健身房 city_demand=0.901 方向=上行 cost=0.8 score=112.66",
   "demand_note": "无源 baseline；用 score 与其均值做基准偏离",
   "algo": "risk_grade",
   "round": "r3",
   "call": "POST http://127.0.0.1:8983/algo/risk_grade",
   "params": {
    "risk_current_value": 150.0,
    "risk_baseline_value": 120.22,
    "risk_trend": 0.5,
    "risk_indicator_name": "重庆健身房geo缺口分"
   },
   "contrast_params": {
    "risk_current_value": 40.0,
    "risk_baseline_value": 120.22,
    "risk_trend": -0.5,
    "risk_indicator_name": "重庆健身房geo缺口分"
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"module_id\": \"risk_grade\", \"name\": \"风险分级算法\", \"indicator_name\": \"重庆健身房geo缺口分\", \"current_value\": 150.0, \"baseline_value\": 120.22, \"deviation_ratio\": 1.2477, \"deviation_pct\": 24.77, \"trend\": 0.5, \"trend_desc\": \"改善↑\", \"trend_emoji\": \"📈\", \"level\": \"稳定\", \"emoji\": \"🟢\", \"level_score\": 1, \"action\": \"维持现状，正常监测\", \"batch_results\": [], \"batch_summary\": \"\", \"summary\": \"🟢 稳定 | 重庆健身房geo缺口分当前150.0，基准120.22，偏离24.8% 📈 改善↑ | 维持现状，正常监测\"}",
   "contrast_excerpt": "{\"module_id\": \"risk_grade\", \"name\": \"风险分级算法\", \"indicator_name\": \"重庆健身房geo缺口分\", \"current_value\": 40.0, \"baseline_value\": 120.22, \"deviation_ratio\": 0.3327, \"deviation_pct\": -66.73, \"trend\": -0.5, \"trend_desc\": \"恶化↓\", \"trend_emoji\": \"📉\", \"level\": \"高危\", \"emoji\": \"🔴\", \"level_score\": 4, \"action\": \"立即处理，紧急干预\", \"batch_results\": [], \"batch_summary\": \"\", \"summary\": \"🔴 高危 | 重庆健身房geo缺口分当前40.0，基准120.22，偏离-66.7% 📉 恶化↓ | 立即处理，紧急干预\"}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "偏离率 × 趋势方向 → 四级。趋势 +0.5 → 改善↑/稳定；-0.5 → 恶化↓/高危。GEO 台账已有 score 与方向，属可立即接。"
  },
  {
   "demand_id": "D4",
   "demand_name": "重庆·健身房 风险等级",
   "demand_src": "geo/gap_top50.csv rank44 重庆 健身房 city_demand=0.901 方向=上行 cost=0.8 score=112.66",
   "demand_note": "无源 baseline；用 score 与其均值做基准偏离",
   "algo": "water_pool",
   "round": "r3",
   "call": "POST http://127.0.0.1:8983/algo/water_pool",
   "params": {
    "water_pool_total_income": 200000,
    "water_pool_total_expense": 50000,
    "water_pool_capacity": 3000000,
    "water_pool_fullness": 0.8,
    "water_pool_compound_rate": 0.15,
    "water_pool_initial_asset": 2000000
   },
   "contrast_params": {
    "water_pool_total_income": 1000,
    "water_pool_total_expense": 90000,
    "water_pool_capacity": 10000,
    "water_pool_fullness": 0.01,
    "water_pool_compound_rate": 0.0,
    "water_pool_initial_asset": 1000
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"module_id\": \"water_pool_module\", \"name\": \"水池理论\", \"total_income\": 200000, \"total_expense\": 50000, \"net_flow\": 150000, \"saving_rate\": 75.0, \"pool_capacity\": 3000000, \"pool_fullness\": 0.8, \"coverage_months\": 60.0, \"compound_rate\": 0.15, \"compound_effect\": 3000.0, \"asset_growth\": 153000.0, \"inlet_quality\": \"充足\", \"outlet_quality\": \"健康\", \"quadrant\": \"有隐患\", \"quadrant_emoji\": \"🟡\", \"strategy\": \"查出水口堵漏\", \"risk_level\": \"中低风险\", \"trend\": \"增长↑\", \"trend_emoji\": \"📈\", \"inlet_count\": 0, \"outlet_count\": 0, \"summary\": \"🟡 有隐患 | 收入200000-支出50000=净流150000 | 结余率75% | 📈增长↑ | 查出水口堵漏\"}",
   "contrast_excerpt": "{\"module_id\": \"water_pool_module\", \"name\": \"水池理论\", \"total_income\": 1000, \"total_expense\": 90000, \"net_flow\": -89000, \"saving_rate\": 0, \"pool_capacity\": 10000, \"pool_fullness\": 0.01, \"coverage_months\": 0.1, \"compound_rate\": 0.0, \"compound_effect\": 0.0, \"asset_growth\": -89000.0, \"inlet_quality\": \"不足\", \"outlet_quality\": \"漏血\", \"quadrant\": \"危急\", \"quadrant_emoji\": \"🔴\", \"strategy\": \"紧急止血+重建\", \"risk_level\": \"极高风险\", \"trend\": \"缩减↓\", \"trend_emoji\": \"📉\", \"inlet_count\": 0, \"outlet_count\": 0, \"summary\": \"🔴 危急 | 收入1000-支出90000=净流-89000 | 结余率0% | 📉缩减↓ | 紧急止血+重建\"}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "带容量/满度/复利率 → 覆盖月数 60、资产增长 153000、入水「充足」、出水「健康」。与 pool_theory_eval 是同一套内核的数据版，选一个即可。"
  },
  {
   "demand_id": "D5",
   "demand_name": "民族服饰AI换装 出海首测$200",
   "demand_src": "ads_scout §5 产品1 infogap.store/dress 首笔$200 看 CPA<=$4 止损 CPM>$40或CPA>$8",
   "demand_note": "渠道单价来自报告实测自述（rd:p:krm9js CPM $15-25 / CPC<$1）",
   "algo": "funnel_evaluate",
   "round": "r2",
   "call": "POST http://127.0.0.1:8983/algo/funnel_evaluate",
   "params": {
    "data": {
     "platform": "Meta",
     "title": "AI民族服饰换装 冷启动",
     "CTR": 0.0153,
     "展现量增速": 0.35,
     "完读率": 0.35,
     "停留时长占比": 0.4,
     "收藏率": 0.05,
     "评论率": 0.01,
     "点赞率": 0.02,
     "私信咨询率": 0.008,
     "加群率": 0.004,
     "关注率": 0.01,
     "转发率": 0.005,
     "裂变比": 0.1
    }
   },
   "contrast_params": {
    "data": {
     "platform": "Meta",
     "title": "AI民族服饰换装 冷启动(健康对照)",
     "CTR": 0.06,
     "展现量增速": 0.5,
     "完读率": 0.7,
     "停留时长占比": 0.9,
     "收藏率": 0.08,
     "评论率": 0.04,
     "点赞率": 0.1,
     "私信咨询率": 0.03,
     "加群率": 0.04,
     "关注率": 0.06,
     "转发率": 0.04,
     "裂变比": 0.6
    }
   },
   "ok": true,
   "ms": 28,
   "output_excerpt": "{\"platform\": \"Meta\", \"title\": \"AI民族服饰换装 冷启动\", \"layers\": {\"entry\": {\"name\": \"入口层\", \"desc\": \"关键词击中意图\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"CTR\", \"status\": \"🔴迷路\", \"value\": 0.0153}, {\"name\": \"展现量增速\", \"status\": \"🟢超线\", \"value\": 0.35}]}, \"engagement\": {\"name\": \"承接层\", \"desc\": \"内容接住流量\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"完读率\", \"status\": \"🟡偏线\", \"value\": 0.35}, {\"name\": \"停留时长占比\", \"status\": \"🔴迷路\", \"value\": 0.4}]}, \"interaction\": {\"name\": \"互动层\", \"desc\": \"产生动作意愿\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"收藏率\", \"status\": \"🟢超线\", \"value\": 0.05}, {\"name\": \"评论率\", \"status\": \"🟡偏线\", \"value\": 0.01}, {\"name\": \"点赞率\", \"status\": \"🔴迷路\", \"value\": 0.02}]}, \"conversion\": {\"name\": \"转化层\", \"desc\": \"从看到找你\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"私信咨询率\", \"status\": \"🟡偏线\", \"value\": 0.008}, {\"name\": \"加群率\", \"status\": \"🔴迷路\", \"value\": 0.004}, {\"name\": \"关注率\", \"status\": \"🔴迷路\", \"value\": 0.01}]}, \"spread\": {\"name\": \"传播层\", \"desc\": \"用户替你推\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"转发率\", \"status\": \"🔴迷路\", \"value\": 0.005}, {\"name\": \"裂变比\", \"status\": \"🔴迷路\", \"value\": 0.1}]}}, \"faults\": []}",
   "contrast_excerpt": "{\"platform\": \"Meta\", \"title\": \"AI民族服饰换装 冷启动(健康对照)\", \"layers\": {\"entry\": {\"name\": \"入口层\", \"desc\": \"关键词击中意图\", \"color\": \"🟢超线\", \"indicators\": [{\"name\": \"CTR\", \"status\": \"🟢超线\", \"value\": 0.06}, {\"name\": \"展现量增速\", \"status\": \"🟢超线\", \"value\": 0.5}]}, \"engagement\": {\"name\": \"承接层\", \"desc\": \"内容接住流量\", \"color\": \"🟢超线\", \"indicators\": [{\"name\": \"完读率\", \"status\": \"🟢超线\", \"value\": 0.7}, {\"name\": \"停留时长占比\", \"status\": \"🟢超线\", \"value\": 0.9}]}, \"interaction\": {\"name\": \"互动层\", \"desc\": \"产生动作意愿\", \"color\": \"🟢超线\", \"indicators\": [{\"name\": \"收藏率\", \"status\": \"🟢超线\", \"value\": 0.08}, {\"name\": \"评论率\", \"status\": \"🟢超线\", \"value\": 0.04}, {\"n",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "必须把数据放进 data dict，且指标名要逐字等于 CTR/展现量增速/完读率/停留时长占比/收藏率/评论率/点赞率/私信咨询率/加群率/关注率/转发率/裂变比——否则 platform 显示「未知」、全部「无数据」。改对后输出五层判色 + 断层定位：坏数据组五层全 🔴迷路，健康对照全 🟢超线。"
  },
  {
   "demand_id": "D5",
   "demand_name": "民族服饰AI换装 出海首测$200",
   "demand_src": "ads_scout §5 产品1 infogap.store/dress 首笔$200 看 CPA<=$4 止损 CPM>$40或CPA>$8",
   "demand_note": "渠道单价来自报告实测自述（rd:p:krm9js CPM $15-25 / CPC<$1）",
   "algo": "traffic_target",
   "round": "r3",
   "call": "POST http://127.0.0.1:8983/algo/traffic_target",
   "params": {
    "industry": "民族服饰AI换装/AI portrait",
    "audience": "US/CA/GB/AU + BR/MX, 18-34 女性",
    "existing_keywords": [
     "ai dress up",
     "traditional dress photo"
    ],
    "platform_results": {
     "ai dress up": {
      "Google": {
       "credibility": 0.8,
       "verified": true,
       "count": 115
      },
      "Bing": {
       "credibility": 0.75,
       "verified": true,
       "count": 169
      },
      "Reddit": {
       "credibility": 0.6,
       "verified": true,
       "count": 1200
      },
      "知乎": {
       "credibility": 0.5,
       "verified": false,
       "count": 30
      }
     },
     "traditional dress photo": {
      "Google": {
       "credibility": 0.7,
       "verified": true,
       "count": 88
      },
      "小红书": {
       "credibility": 0.65,
       "verified": true,
       "count": 400
      },
      "百度": {
       "credibility": 0.4,
       "verified": false,
       "count": 12
      }
     }
    }
   },
   "contrast_params": {
    "industry": "民族服饰AI换装/AI portrait",
    "audience": "x",
    "existing_keywords": [
     "ai dress up",
     "traditional dress photo"
    ],
    "platform_results": {
     "ai dress up": {
      "Google": {
       "credibility": 0.1,
       "verified": false,
       "count": 0
      }
     },
     "traditional dress photo": {
      "Google": {
       "credibility": 0.1,
       "verified": false,
       "count": 0
      }
     }
    }
   },
   "ok": true,
   "ms": 21,
   "output_excerpt": "{\"industry\": \"民族服饰AI换装/AI portrait\", \"audience\": \"US/CA/GB/AU + BR/MX, 18-34 女性\", \"target_keywords\": [{\"keyword\": \"ai dress up\", \"platforms\": {\"Google\": {\"verified\": true, \"credibility\": 0.8, \"count\": 115}, \"Bing\": {\"verified\": true, \"credibility\": 0.75, \"count\": 169}, \"Reddit\": {\"verified\": true, \"credibility\": 0.6, \"count\": 1200}, \"知乎\": {\"verified\": false, \"credibility\": 0.5, \"count\": 30}}, \"cross_validation_count\": 3, \"weighted_rate\": 0.19, \"priority\": \"P3\"}, {\"keyword\": \"traditional dress photo\", \"platforms\": {\"Google\": {\"verified\": true, \"credibility\": 0.7, \"count\": 88}, \"小红书\": {\"verified\": true, \"credibility\": 0.65, \"count\": 400}, \"百度\": {\"verified\": false, \"credibility\": 0.4, \"count\": 12}}, \"cross_validation_count\": 2, \"weighted_rate\": 0.08, \"priority\": \"P3\"}], \"eliminated_keywords\": [], \"platform_info\": {\"知乎\": {\"name\": \"知乎\", \"type\": \"开放\", \"verify\": \"site:搜索\"}, \"小红书\": {\"name\": \"小红书\", \"type\": \"封闭花园\", \"verify\": \"App内实搜\"}, \"头条\": {\"name\": \"头条\", \"type\": \"半开放\", \"verify\": \"site:搜索\"}, \"抖音\": {\"name\": \"抖音\", \"type\": \"封闭花园\", \"verify\": \"App内实搜\"}, \"快手\": {\"name\": \"快手\", \"type\": \"封闭花园\", \"verify\": \"App内实搜\"}, \"B站\": {\"name\": \"B站\", \"type\": \"半开放\", \"verify\": \"site:搜索\"}, \"微信搜一搜\": {\"name\": \"微信搜一搜\", \"",
   "contrast_excerpt": "{\"industry\": \"民族服饰AI换装/AI portrait\", \"audience\": \"x\", \"target_keywords\": [{\"keyword\": \"ai dress up\", \"platforms\": {\"Google\": {\"verified\": false, \"credibility\": 0.1, \"count\": 0}}, \"cross_validation_count\": 0, \"weighted_rate\": 0.0, \"priority\": \"P3\"}, {\"keyword\": \"traditional dress photo\", \"platforms\": {\"Google\": {\"verified\": false, \"credibility\": 0.1, \"count\": 0}}, \"cross_validation_count\": 0, \"weighted_rate\": 0.0, \"priority\": \"P3\"}], \"eliminated_keywords\": [], \"platform_info\": {\"知乎\": {\"name\": \"知乎\", \"type\": \"开放\", \"verify\": \"site:搜索\"}, \"小红书\": {\"name\": \"小红书\", \"type\": \"封闭花园\", \"verify\": \"App内实搜\"}, \"",
   "param_sensitive": true,
   "verdict": "勉强",
   "reason": "必须传 platform_results[关键词][平台名]={credibility,verified,count} 才算得出交叉验证（实测 cross_validation_count 3/2、weighted_rate 0.19/0.08）。但 priority 恒为 P3、eliminated_keywords 恒空，输出接不到「选词」下一步；且平台词表是知乎/小红书/头条/抖音等国内 8 平台，对出海英文词不对口。"
  },
  {
   "demand_id": "D6",
   "demand_name": "多语言方法论内容 newsletter $200",
   "demand_src": "ads_scout §5 产品2 infogap.store/slices 首笔$200 投8-12个垂直newsletter 看注册成本<$2",
   "demand_note": "newsletter CPM $250 来自 rd:p:1g7xuy3；$2000/20+刊 来自 rd:p:199rxrm",
   "algo": "funnel_evaluate",
   "round": "r2",
   "call": "POST http://127.0.0.1:8983/algo/funnel_evaluate",
   "params": {
    "data": {
     "platform": "Newsletter",
     "title": "8-12个垂直刊首测",
     "CTR": 0.0153,
     "展现量增速": 0.35,
     "完读率": 0.35,
     "停留时长占比": 0.4,
     "收藏率": 0.05,
     "评论率": 0.01,
     "点赞率": 0.02,
     "私信咨询率": 0.008,
     "加群率": 0.004,
     "关注率": 0.01,
     "转发率": 0.005,
     "裂变比": 0.1
    }
   },
   "contrast_params": {
    "data": {
     "platform": "Newsletter",
     "title": "8-12个垂直刊首测(对照)",
     "CTR": 0.06,
     "展现量增速": 0.5,
     "完读率": 0.7,
     "停留时长占比": 0.9,
     "收藏率": 0.08,
     "评论率": 0.04,
     "点赞率": 0.1,
     "私信咨询率": 0.03,
     "加群率": 0.04,
     "关注率": 0.06,
     "转发率": 0.04,
     "裂变比": 0.6
    }
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"platform\": \"Newsletter\", \"title\": \"8-12个垂直刊首测\", \"layers\": {\"entry\": {\"name\": \"入口层\", \"desc\": \"关键词击中意图\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"CTR\", \"status\": \"🔴迷路\", \"value\": 0.0153}, {\"name\": \"展现量增速\", \"status\": \"🟢超线\", \"value\": 0.35}]}, \"engagement\": {\"name\": \"承接层\", \"desc\": \"内容接住流量\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"完读率\", \"status\": \"🟡偏线\", \"value\": 0.35}, {\"name\": \"停留时长占比\", \"status\": \"🔴迷路\", \"value\": 0.4}]}, \"interaction\": {\"name\": \"互动层\", \"desc\": \"产生动作意愿\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"收藏率\", \"status\": \"🟢超线\", \"value\": 0.05}, {\"name\": \"评论率\", \"status\": \"🟡偏线\", \"value\": 0.01}, {\"name\": \"点赞率\", \"status\": \"🔴迷路\", \"value\": 0.02}]}, \"conversion\": {\"name\": \"转化层\", \"desc\": \"从看到找你\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"私信咨询率\", \"status\": \"🟡偏线\", \"value\": 0.008}, {\"name\": \"加群率\", \"status\": \"🔴迷路\", \"value\": 0.004}, {\"name\": \"关注率\", \"status\": \"🔴迷路\", \"value\": 0.01}]}, \"spread\": {\"name\": \"传播层\", \"desc\": \"用户替你推\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"转发率\", \"status\": \"🔴迷路\", \"value\": 0.005}, {\"name\": \"裂变比\", \"status\": \"🔴迷路\", \"value\": 0.1}]}}, \"faults\": []}",
   "contrast_excerpt": "{\"platform\": \"Newsletter\", \"title\": \"8-12个垂直刊首测(对照)\", \"layers\": {\"entry\": {\"name\": \"入口层\", \"desc\": \"关键词击中意图\", \"color\": \"🟢超线\", \"indicators\": [{\"name\": \"CTR\", \"status\": \"🟢超线\", \"value\": 0.06}, {\"name\": \"展现量增速\", \"status\": \"🟢超线\", \"value\": 0.5}]}, \"engagement\": {\"name\": \"承接层\", \"desc\": \"内容接住流量\", \"color\": \"🟢超线\", \"indicators\": [{\"name\": \"完读率\", \"status\": \"🟢超线\", \"value\": 0.7}, {\"name\": \"停留时长占比\", \"status\": \"🟢超线\", \"value\": 0.9}]}, \"interaction\": {\"name\": \"互动层\", \"desc\": \"产生动作意愿\", \"color\": \"🟢超线\", \"indicators\": [{\"name\": \"收藏率\", \"status\": \"🟢超线\", \"value\": 0.08}, {\"name\": \"评论率\", \"status\": \"🟢超线\", \"value\": 0.04}, {",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "同 D5；Newsletter 场景五层判色可用，但它是「内容漏斗」口径，不是「刊位 CPM」口径——别拿它替 newsletter 的 $250 CPM 算账。"
  },
  {
   "demand_id": "D6",
   "demand_name": "多语言方法论内容 newsletter $200",
   "demand_src": "ads_scout §5 产品2 infogap.store/slices 首笔$200 投8-12个垂直newsletter 看注册成本<$2",
   "demand_note": "newsletter CPM $250 来自 rd:p:1g7xuy3；$2000/20+刊 来自 rd:p:199rxrm",
   "algo": "traffic_target",
   "round": "r2",
   "call": "POST http://127.0.0.1:8983/algo/traffic_target",
   "params": {
    "industry": "多语言方法论内容/知识付费",
    "audience": "美/英/德/西/墨/沙特/巴西 独立开发者",
    "existing_keywords": [
     "life chain method",
     "business chain diagnosis"
    ],
    "platform_results": {
     "google": {
      "name": "Google",
      "count": 115,
      "verified": true,
      "credibility": 0.8
     },
     "bing": {
      "name": "Bing",
      "count": 169,
      "verified": true,
      "credibility": 0.75
     },
     "reddit": {
      "name": "Reddit",
      "count": 1200,
      "verified": true,
      "credibility": 0.6
     },
     "newsletter": {
      "name": "Newsletter",
      "count": 20,
      "verified": true,
      "credibility": 0.55
     }
    },
    "credibility": 0.7,
    "verified": true,
    "count": 284
   },
   "contrast_params": {
    "industry": "多语言方法论内容/知识付费",
    "audience": "x",
    "existing_keywords": [
     "life chain method"
    ],
    "platform_results": {
     "google": {
      "name": "Google",
      "count": 1,
      "verified": false,
      "credibility": 0.05
     }
    },
    "credibility": 0.05,
    "verified": false,
    "count": 1
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"industry\": \"多语言方法论内容/知识付费\", \"audience\": \"美/英/德/西/墨/沙特/巴西 独立开发者\", \"target_keywords\": [{\"keyword\": \"life chain method\", \"platforms\": {}, \"cross_validation_count\": 0, \"weighted_rate\": 0.0, \"priority\": \"P3\"}, {\"keyword\": \"business chain diagnosis\", \"platforms\": {}, \"cross_validation_count\": 0, \"weighted_rate\": 0.0, \"priority\": \"P3\"}], \"eliminated_keywords\": [], \"platform_info\": {\"知乎\": {\"name\": \"知乎\", \"type\": \"开放\", \"verify\": \"site:搜索\"}, \"小红书\": {\"name\": \"小红书\", \"type\": \"封闭花园\", \"verify\": \"App内实搜\"}, \"头条\": {\"name\": \"头条\", \"type\": \"半开放\", \"verify\": \"site:搜索\"}, \"抖音\": {\"name\": \"抖音\", \"type\": \"封闭花园\", \"verify\": \"App内实搜\"}, \"快手\": {\"name\": \"快手\", \"type\": \"封闭花园\", \"verify\": \"App内实搜\"}, \"B站\": {\"name\": \"B站\", \"type\": \"半开放\", \"verify\": \"site:搜索\"}, \"微信搜一搜\": {\"name\": \"微信搜一搜\", \"type\": \"半封闭\", \"verify\": \"site:搜索\"}, \"百度\": {\"name\": \"百度\", \"type\": \"开放\", \"verify\": \"通用搜索\"}}}",
   "contrast_excerpt": "{\"industry\": \"多语言方法论内容/知识付费\", \"audience\": \"x\", \"target_keywords\": [{\"keyword\": \"life chain method\", \"platforms\": {}, \"cross_validation_count\": 0, \"weighted_rate\": 0.0, \"priority\": \"P3\"}], \"eliminated_keywords\": [], \"platform_info\": {\"知乎\": {\"name\": \"知乎\", \"type\": \"开放\", \"verify\": \"site:搜索\"}, \"小红书\": {\"name\": \"小红书\", \"type\": \"封闭花园\", \"verify\": \"App内实搜\"}, \"头条\": {\"name\": \"头条\", \"type\": \"半开放\", \"verify\": \"site:搜索\"}, \"抖音\": {\"name\": \"抖音\", \"type\": \"封闭花园\", \"verify\": \"App内实搜\"}, \"快手\": {\"name\": \"快手\", \"type\": \"封闭花园\", \"verify\": \"App内实搜\"}, \"B站\": {\"name\": \"B站\", \"type\": \"半开放\", \"verify\": \"site:搜索\"}, \"微信搜一搜\": {\"name\": ",
   "param_sensitive": true,
   "verdict": "勉强",
   "reason": "同 D5：能算交叉验证，但 priority 恒 P3、无淘汰，且国内平台词表与海外 newsletter 场景错配。"
  },
  {
   "demand_id": "D7",
   "demand_name": "生命链平台 海外华人 Reddit冷启动$150",
   "demand_src": "ads_scout §5 产品4 首笔$150 = $50 Reddit($5-10/天×7) + $100 两个newsletter 看留资成本<=$3",
   "demand_note": "Reddit 实测：$700→77000曝光/180点击/0注册(rd:p:1vmp84p)；CPC$0.19 CPM$9.65 784点击1线索(rd:p:1253zfi)",
   "algo": "decay",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/decay",
   "params": {
    "initial_users": 180,
    "avg_invitations": 0.23,
    "max_depth": 8,
    "decay_rate": 0.35,
    "spread_data": {
     "initial_users": 180,
     "avg_invitations": 0.23,
     "decay_rate": 0.35
    }
   },
   "contrast_params": {
    "initial_users": 180,
    "avg_invitations": 2.5,
    "max_depth": 8,
    "decay_rate": 0.05
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"layers\": [{\"depth\": 1, \"users\": 180.0, \"decay_factor\": 1.0, \"cumulative\": 180.0}, {\"depth\": 2, \"users\": 14.35, \"decay_factor\": 0.3465, \"cumulative\": 194.35}, {\"depth\": 3, \"users\": 1.13, \"decay_factor\": 0.3431, \"cumulative\": 195.48}, {\"depth\": 4, \"users\": 0.09, \"decay_factor\": 0.3397, \"cumulative\": 195.57}], \"total_spread\": 195, \"decay_curve\": \"衰减显著，末层仅剩0.1%\", \"max_depth\": 4}",
   "contrast_excerpt": "{\"layers\": [{\"depth\": 1, \"users\": 180.0, \"decay_factor\": 1.0, \"cumulative\": 180.0}, {\"depth\": 2, \"users\": 22.28, \"decay_factor\": 0.0495, \"cumulative\": 202.28}, {\"depth\": 3, \"users\": 2.73, \"decay_factor\": 0.049, \"cumulative\": 205.01}, {\"depth\": 4, \"users\": 0.33, \"decay_factor\": 0.0485, \"cumulative\": 205.34}], \"total_spread\": 205, \"decay_curve\": \"衰减显著，末层仅剩0.2%\", \"max_depth\": 4}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "同 input 下与 water_flow 前几代口径一致，输出「衰减显著，末层仅剩 0.1%」。注意它与 water_flow 高度重复，两个都上等于重复计数。"
  },
  {
   "demand_id": "D7",
   "demand_name": "生命链平台 海外华人 Reddit冷启动$150",
   "demand_src": "ads_scout §5 产品4 首笔$150 = $50 Reddit($5-10/天×7) + $100 两个newsletter 看留资成本<=$3",
   "demand_note": "Reddit 实测：$700→77000曝光/180点击/0注册(rd:p:1vmp84p)；CPC$0.19 CPM$9.65 784点击1线索(rd:p:1253zfi)",
   "algo": "k_factor",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/k_factor",
   "params": {
    "invitations": 0.23,
    "conversions": 0.0013,
    "spread_data": {
     "invites_per_user": 0.23,
     "conversion_rate": 0.0013
    }
   },
   "contrast_params": {
    "invitations": 3.0,
    "conversions": 0.4,
    "spread_data": {
     "invites_per_user": 3.0,
     "conversion_rate": 0.4
    }
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"k_factor\": 0.0013, \"is_viral\": false, \"conversion_rate\": 0.0057, \"invitations\": 0.23, \"conversions\": 0.0013, \"phase\": \"seed\", \"distance_to_viral\": 0.9987, \"suggestions\": [\"K值过低（种子期），优先提升邀请率和转化率\"]}",
   "contrast_excerpt": "{\"k_factor\": 0.4, \"is_viral\": false, \"conversion_rate\": 0.1333, \"invitations\": 3.0, \"conversions\": 0.4, \"phase\": \"growth\", \"distance_to_viral\": 0.6, \"suggestions\": [\"K值增长期(K=0.40)，距病毒传播还有空间\"]}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "invites_per_user × conversion_rate → K=0.0013，phase=seed，distance_to_viral=0.9987。一句话结论「K 值过低，优先提升邀请率和转化率」。"
  },
  {
   "demand_id": "D7",
   "demand_name": "生命链平台 海外华人 Reddit冷启动$150",
   "demand_src": "ads_scout §5 产品4 首笔$150 = $50 Reddit($5-10/天×7) + $100 两个newsletter 看留资成本<=$3",
   "demand_note": "Reddit 实测：$700→77000曝光/180点击/0注册(rd:p:1vmp84p)；CPC$0.19 CPM$9.65 784点击1线索(rd:p:1253zfi)",
   "algo": "water_flow",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/water_flow",
   "params": {
    "initial_users": 180,
    "invites_per_user": 0.23,
    "conversion_rate": 0.0013,
    "decay_rate": 0.35,
    "max_generations": 8,
    "avg_value_per_user": 3.0,
    "propagation_input": {
     "initial_users": 180,
     "invites_per_user": 0.23,
     "conversion_rate": 0.0013,
     "decay_rate": 0.35
    }
   },
   "contrast_params": {
    "initial_users": 180,
    "invites_per_user": 2.5,
    "conversion_rate": 0.2,
    "decay_rate": 0.05,
    "max_generations": 8,
    "avg_value_per_user": 3.0
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"k_factor\": 0.0003, \"phase\": \"seed\", \"is_viral\": false, \"generations\": [{\"generation\": 1, \"users\": 180.0, \"decay_factor\": 1.0, \"cumulative\": 180.0}, {\"generation\": 2, \"users\": 14.35, \"decay_factor\": 0.3465, \"cumulative\": 194.35}, {\"generation\": 3, \"users\": 1.13, \"decay_factor\": 0.3431, \"cumulative\": 195.48}, {\"generation\": 4, \"users\": 0.09, \"decay_factor\": 0.3397, \"cumulative\": 195.57}], \"total_reach\": 195.6, \"peak_generation\": 1, \"peak_users\": 180.0, \"network_value\": 2.723, \"critical_progress_pct\": 19.5, \"health_score\": 39.9, \"health_grade\": \"D\"}",
   "contrast_excerpt": "{\"k_factor\": 0.5, \"phase\": \"growth\", \"is_viral\": false, \"generations\": [{\"generation\": 1, \"users\": 180.0, \"decay_factor\": 1.0, \"cumulative\": 180.0}, {\"generation\": 2, \"users\": 22.28, \"decay_factor\": 0.0495, \"cumulative\": 202.28}, {\"generation\": 3, \"users\": 2.73, \"decay_factor\": 0.049, \"cumulative\": 205.01}, {\"generation\": 4, \"users\": 0.33, \"decay_factor\": 0.0485, \"cumulative\": 205.34}], \"total_reach\": 205.3, \"peak_generation\": 1, \"peak_users\": 180.0, \"network_value\": 2.9352, \"critical_progress_pct\": 20.5, \"health_score\": 50.1, \"health_grade\": \"D\"}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "逐代用户数 + 总触达 195.6 + 网络价值 2.723 + 健康度 D。能把「$700 买 180 点击 0 注册」翻译成传播曲线。"
  },
  {
   "demand_id": "D8",
   "demand_name": "TripCurve 行程规划 Google长尾$150",
   "demand_src": "ads_scout §5 产品5 首笔$150($20/天×7 长尾词)+$200微网红 看每次生成<=$1.5 生成→注册>=8% 止损CPC>$3",
   "demand_note": "Google CPC $0.4-2.6 来自报告 §2",
   "algo": "funnel_evaluate",
   "round": "r2",
   "call": "POST http://127.0.0.1:8983/algo/funnel_evaluate",
   "params": {
    "data": {
     "platform": "Google",
     "title": "itinerary generator 长尾词",
     "CTR": 0.0153,
     "展现量增速": 0.35,
     "完读率": 0.35,
     "停留时长占比": 0.4,
     "收藏率": 0.05,
     "评论率": 0.01,
     "点赞率": 0.02,
     "私信咨询率": 0.008,
     "加群率": 0.004,
     "关注率": 0.01,
     "转发率": 0.005,
     "裂变比": 0.1
    }
   },
   "contrast_params": {
    "data": {
     "platform": "Google",
     "title": "itinerary generator 长尾词(对照)",
     "CTR": 0.06,
     "展现量增速": 0.5,
     "完读率": 0.7,
     "停留时长占比": 0.9,
     "收藏率": 0.08,
     "评论率": 0.04,
     "点赞率": 0.1,
     "私信咨询率": 0.03,
     "加群率": 0.04,
     "关注率": 0.06,
     "转发率": 0.04,
     "裂变比": 0.6
    }
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"platform\": \"Google\", \"title\": \"itinerary generator 长尾词\", \"layers\": {\"entry\": {\"name\": \"入口层\", \"desc\": \"关键词击中意图\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"CTR\", \"status\": \"🔴迷路\", \"value\": 0.0153}, {\"name\": \"展现量增速\", \"status\": \"🟢超线\", \"value\": 0.35}]}, \"engagement\": {\"name\": \"承接层\", \"desc\": \"内容接住流量\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"完读率\", \"status\": \"🟡偏线\", \"value\": 0.35}, {\"name\": \"停留时长占比\", \"status\": \"🔴迷路\", \"value\": 0.4}]}, \"interaction\": {\"name\": \"互动层\", \"desc\": \"产生动作意愿\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"收藏率\", \"status\": \"🟢超线\", \"value\": 0.05}, {\"name\": \"评论率\", \"status\": \"🟡偏线\", \"value\": 0.01}, {\"name\": \"点赞率\", \"status\": \"🔴迷路\", \"value\": 0.02}]}, \"conversion\": {\"name\": \"转化层\", \"desc\": \"从看到找你\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"私信咨询率\", \"status\": \"🟡偏线\", \"value\": 0.008}, {\"name\": \"加群率\", \"status\": \"🔴迷路\", \"value\": 0.004}, {\"name\": \"关注率\", \"status\": \"🔴迷路\", \"value\": 0.01}]}, \"spread\": {\"name\": \"传播层\", \"desc\": \"用户替你推\", \"color\": \"🔴迷路\", \"indicators\": [{\"name\": \"转发率\", \"status\": \"🔴迷路\", \"value\": 0.005}, {\"name\": \"裂变比\", \"status\": \"🔴迷路\", \"value\": 0.1}]}}, \"faults\": []}",
   "contrast_excerpt": "{\"platform\": \"Google\", \"title\": \"itinerary generator 长尾词(对照)\", \"layers\": {\"entry\": {\"name\": \"入口层\", \"desc\": \"关键词击中意图\", \"color\": \"🟢超线\", \"indicators\": [{\"name\": \"CTR\", \"status\": \"🟢超线\", \"value\": 0.06}, {\"name\": \"展现量增速\", \"status\": \"🟢超线\", \"value\": 0.5}]}, \"engagement\": {\"name\": \"承接层\", \"desc\": \"内容接住流量\", \"color\": \"🟢超线\", \"indicators\": [{\"name\": \"完读率\", \"status\": \"🟢超线\", \"value\": 0.7}, {\"name\": \"停留时长占比\", \"status\": \"🟢超线\", \"value\": 0.9}]}, \"interaction\": {\"name\": \"互动层\", \"desc\": \"产生动作意愿\", \"color\": \"🟢超线\", \"indicators\": [{\"name\": \"收藏率\", \"status\": \"🟢超线\", \"value\": 0.08}, {\"name\": \"评论率\", \"status\": \"🟢超线\", \"value\":",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "Google 长尾场景判色可用（三组平台三组判色，互不串味），可直接当投放日报的判色器。"
  },
  {
   "demand_id": "D8",
   "demand_name": "TripCurve 行程规划 Google长尾$150",
   "demand_src": "ads_scout §5 产品5 首笔$150($20/天×7 长尾词)+$200微网红 看每次生成<=$1.5 生成→注册>=8% 止损CPC>$3",
   "demand_note": "Google CPC $0.4-2.6 来自报告 §2",
   "algo": "network_effect",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/network_effect",
   "params": {
    "total_users": 700,
    "connections": 1.4,
    "avg_value_per_user": 1.5,
    "alpha": 1.1,
    "beta": 0.9,
    "c": 1.0,
    "n_critical": 1000
   },
   "contrast_params": {
    "total_users": 50000,
    "connections": 12,
    "avg_value_per_user": 25,
    "alpha": 1.1,
    "beta": 0.9,
    "c": 1.0,
    "n_critical": 1000
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"network_value\": 1347.7454, \"metcalfe_value\": 367500.0, \"phase\": \"pre_critical\", \"density\": 6e-06, \"distance_to_critical\": 300, \"critical_progress_pct\": 70.0, \"total_users\": 700, \"n_critical\": 1000}",
   "contrast_excerpt": "{\"network_value\": 16946.2264, \"metcalfe_value\": 31250000000.0, \"phase\": \"post_critical\", \"density\": 0.0, \"distance_to_critical\": 0, \"critical_progress_pct\": 100.0, \"total_users\": 50000, \"n_critical\": 1000}",
   "param_sensitive": true,
   "verdict": "勉强",
   "reason": "能算 V=1347.75、临界进度 70%；但 density=6e-06、distance_to_critical 口径未标单位，输出读到人眼会误判。要先补口径说明才能进决策。"
  },
  {
   "demand_id": "D8",
   "demand_name": "TripCurve 行程规划 Google长尾$150",
   "demand_src": "ads_scout §5 产品5 首笔$150($20/天×7 长尾词)+$200微网红 看每次生成<=$1.5 生成→注册>=8% 止损CPC>$3",
   "demand_note": "Google CPC $0.4-2.6 来自报告 §2",
   "algo": "spread_validation",
   "round": "r3",
   "call": "POST http://127.0.0.1:8983/algo/spread_validation",
   "params": {
    "invites_per_user": 2.5,
    "conversion_rate": 0.3,
    "decay_rate": 0.7,
    "spread_data": {
     "invites_per_user": 2.5,
     "conversion_rate": 0.3,
     "decay_rate": 0.7
    }
   },
   "contrast_params": {
    "invites_per_user": 0.2,
    "conversion_rate": 0.001,
    "decay_rate": 0.95
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"score\": 1.0, \"valid_rate\": 1.0, \"fraud_flag\": \"clean\", \"anomalies\": [], \"anomaly_count\": 0}",
   "contrast_excerpt": "{\"score\": 1.0, \"valid_rate\": 1.0, \"fraud_flag\": \"suspect\", \"anomalies\": [\"每人邀请数过低(0.2)，传播动力不足\", \"转化率过低(0%)，产品匹配度差\"], \"anomaly_count\": 2}",
   "param_sensitive": true,
   "verdict": "废",
   "reason": "score 与 valid_rate 恒为 1.0——好输入坏输入都是 1.0，只有 fraud_flag 会从 clean 翻成 suspect。一个「评分」字段永远满分，等于没有。要留只能只用 fraud_flag/anomalies。"
  },
  {
   "demand_id": "D9",
   "demand_name": "分身引擎卖点 → 验证是否真需求（民族服饰AI换装）",
   "demand_src": "commander_engine 分身引擎_算法版_卖点张老师 产出\"卖点\"；本需求=该卖点背后是不是真需求",
   "demand_note": "卖点文案来自引擎产出场景，本机不打开引擎（只读场景定义）",
   "algo": "demand_verify",
   "round": "r2",
   "call": "POST http://127.0.0.1:8983/algo/demand_verify",
   "params": {
    "answers": {
     "q1_self_statement": "分身引擎卖点：一张照片穿遍民族服饰",
     "q2_searched": 2,
     "q3_tried": 1,
     "q4_paid": 0,
     "q5_willing": 2,
     "q6_action": 1
    }
   },
   "contrast_params": {
    "answers": {
     "q1_self_statement": "同左",
     "q2_searched": 0,
     "q3_tried": 0,
     "q4_paid": 0,
     "q5_willing": 0,
     "q6_action": 0
    }
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"q1_self_statement\": \"分身引擎卖点：一张照片穿遍民族服饰\", \"behavior_score\": 55, \"commitment_score\": 10.3, \"total_score\": 65.3, \"level\": \"🟡潜在需求\", \"confidence\": \"中\", \"behavior_details\": [\"✅ 主动搜索过解决方案(+25)\", \"✅ 尝试过其他方案(+30)\", \"❌ 未付过费(0)\", \"✅ 愿意采取具体行动(+10)\"], \"suggestion\": \"引导行动，降低首次尝试门槛\"}",
   "contrast_excerpt": "{\"q1_self_statement\": \"同左\", \"behavior_score\": 0, \"commitment_score\": 0.0, \"total_score\": 0.0, \"level\": \"🔴伪需求\", \"confidence\": \"极低\", \"behavior_details\": [\"❌ 未搜索过(0)\", \"❌ 未尝试过(0)\", \"❌ 未付过费(0)\"], \"suggestion\": \"不投入资源，提供免费内容即可\"}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "卖点场景：搜索+尝试 → 65.3 🟡潜在需求（中），建议「降低首次尝试门槛」；对照不给行为 → 0 分伪需求。同一算法在两个不同需求上给出不同分级，可用。"
  },
  {
   "demand_id": "D9",
   "demand_name": "分身引擎卖点 → 验证是否真需求（民族服饰AI换装）",
   "demand_src": "commander_engine 分身引擎_算法版_卖点张老师 产出\"卖点\"；本需求=该卖点背后是不是真需求",
   "demand_note": "卖点文案来自引擎产出场景，本机不打开引擎（只读场景定义）",
   "algo": "motivation_level",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/motivation_level",
   "params": {
    "referral_count": 3,
    "name": "参与试用的海外华人用户"
   },
   "contrast_params": {
    "referral_count": 0,
    "name": "x"
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"tier\": \"T1\", \"tier_name\": \"利他分享\", \"referral_count\": 3, \"earnings_rate\": 0.7, \"benefits\": [\"70%终身分成\", \"普通推广人标识\", \"糖丸奖励×5/人\"], \"next_tier\": \"T3\", \"progress_to_next\": 0.33, \"next_tier_name\": \"利益驱动\"}",
   "contrast_excerpt": "{\"tier\": \"T0\", \"tier_name\": \"旁观者\", \"referral_count\": 0, \"earnings_rate\": 0.0, \"benefits\": [\"免费看路\", \"基础功能\"], \"next_tier\": \"T1\", \"progress_to_next\": 0.0, \"next_tier_name\": \"利他分享\"}",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "referral_count=3 → T1 利他分享、70% 终身分成、下一档进度 0.33。可直接挂推广后台。"
  },
  {
   "demand_id": "D9",
   "demand_name": "分身引擎卖点 → 验证是否真需求（民族服饰AI换装）",
   "demand_src": "commander_engine 分身引擎_算法版_卖点张老师 产出\"卖点\"；本需求=该卖点背后是不是真需求",
   "demand_note": "卖点文案来自引擎产出场景，本机不打开引擎（只读场景定义）",
   "algo": "value_assessment_v123",
   "round": "r1",
   "call": "POST http://127.0.0.1:8983/algo/value_assessment_v123",
   "params": {
    "product_name": "AI民族服饰换装",
    "product_desc": "上传1张照片生成民族服饰写真；海外晒图/节庆",
    "efficiency_gain": 8,
    "cost_reduction": 9,
    "coverage_scale": 6,
    "substitution_risk": 0.6,
    "security_score": 3,
    "pleasure_score": 9,
    "showcase_score": 10,
    "emotion_duration": 0.4,
    "secondary_market": 2,
    "consensus_strength": 4,
    "value_period": 0.3,
    "scarcity": 4
   },
   "contrast_params": {
    "product_name": "AI民族服饰换装",
    "product_desc": "x",
    "efficiency_gain": 1,
    "cost_reduction": 1,
    "coverage_scale": 1,
    "substitution_risk": 0.95,
    "security_score": 1,
    "pleasure_score": 1,
    "showcase_score": 1,
    "emotion_duration": 0.05,
    "secondary_market": 1,
    "consensus_strength": 1,
    "value_period": 0.05,
    "scarcity": 1
   },
   "ok": true,
   "ms": 0,
   "output_excerpt": "{\"module\": \"真需求价值评估 V1~V3\", \"product_name\": \"AI民族服饰换装\", \"product_desc\": \"上传1张照片生成民族服饰写真；海外晒图/节庆\", \"v1_functional\": {\"score\": 0.91, \"base_score\": 1.0, \"model_bonus\": 0, \"risk_penalty\": 0.09, \"model_identified\": {\"model\": \"infrastructure\", \"name\": \"基础设施模型\", \"confidence\": \"高\"}, \"model_input\": \"\", \"substitution_risk\": 0.6, \"level\": \"🟢极高\", \"eval\": \"功能价值极高\"}, \"v2_emotional\": {\"score\": 0.82, \"base_score\": 1.0, \"duration_factor\": 0.82, \"security\": 1, \"pleasure\": 1, \"showcase\": 1, \"dominant_element\": {\"id\": \"security\", \"name\": \"保障感\", \"score\": 1, \"desc\": \"对抗害怕\"}, \"level\": \"🟢极高\", \"eval\": \"情绪价值极强\"}, \"v3_asset\": {\"score\": 0.8, \"base_score\": 1.0, \"period_factor\": 0.5, \"secondary_market\": 1, \"consensus_strength\": 1, \"scarcity\": 1, \"value_period_years\": 0.3, \"has_market\": true, \"has_consensus\": true, \"level\": \"🟢高\", \"eval\": \"资产价值良好\"}, \"F_score\": 0.91, \"E_score\": 0.82, \"A_score\": 0.8, \"V_total\": 0.845, \"v …[TRUNC]",
   "contrast_excerpt": "{\"module\": \"真需求价值评估 V1~V3\", \"product_name\": \"AI民族服饰换装\", \"product_desc\": \"x\", \"v1_functional\": {\"score\": 0.858, \"base_score\": 1.0, \"model_bonus\": 0, \"risk_penalty\": 0.142, \"model_identified\": {\"model\": \"infrastructure\", \"name\": \"基础设施模型\", \"confidence\": \"高\"}, \"model_input\": \"\", \"substitution_risk\": 0.95, \"level\": \"🟢极高\", \"eval\": \"功能价值极高\"}, \"v2_emotional\": {\"score\": 0.715, \"base_score\": 1.0, \"duration_factor\": 0.715, \"security\": 1, \"pleasure\": 1, \"showcase\": 1, \"dominant_element\": {\"id\": \"security\", \"name\": \"保障感\", \"score\": 1, \"desc\": \"对抗害怕\"}, \"level\": \"🟢高\", \"eval\": \"情绪价值良好\"}, \"v3_asset\": {\"score\": ",
   "param_sensitive": true,
   "verdict": "有用",
   "reason": "AI民族服饰换装：V1 0.91 / V2 0.82 / V3 0.80，识别为「基础设施模型」。与中医馆那组分数接近，说明它对不同品类分辨力有限，但输出结构可直接进产品页。"
  }
 ],
 "first_batch": [
  {
   "algo": "value_assessment_v123",
   "demand": "D2 上海·中医馆 空位是不是真需求 / D9 AI 换装卖点",
   "output_sample": "真需求价值评估 V1~V3 · 值不值得做",
   "hookup": "/internal/geo/ 每行城市×行业加一颗「值不值得做」按钮；产品立项前先跑一次",
   "must_add": "产品名 + 描述 + 功能/情绪/资产三组评分（部分可从 geo 台账换算）",
   "tracking": "dispatcher 目前无访问日志（log_message 被关掉），要先加一行调用计数",
   "value": "高：直接决定「做/不做」，省掉一轮人工评审",
   "entry": "/internal/geo/ 每行「值不值得做」弹窗",
   "entry_status": "已接上线 2026-09-21"
  },
  {
   "algo": "demand_verify",
   "demand": "D2 上海·中医馆 / D9 分身引擎卖点",
   "output_sample": "需求验证6题 · 真需求/伪需求分级",
   "hookup": "投放闸门 + 立项闸门；挂到 /algo53/ 之后由运营手填 5 题",
   "must_add": "必须用 answers 嵌套 dict 传 q1~q6，顶层键不生效",
   "tracking": "同上；另需记录 answer 组合以便复算",
   "value": "高：200 元级投放前的一道闸，避免把钱花在伪需求上",
   "entry": "/internal/hit/gate.html 立项闸门",
   "entry_status": "已接上线 2026-09-21"
  },
  {
   "algo": "funnel_evaluate",
   "demand": "D5/D6/D8 三条渠道首测",
   "output_sample": "5层靶心漏斗评估 · 五层判色 + 断层",
   "hookup": "投放日报 / 周报页；每投一条就填一次",
   "must_add": "data dict 内必须含 platform/title + 12 个逐字指标名，缺一个即「无数据」",
   "tracking": "同上；建议记录每次判色结果形成时间序列",
   "value": "高：把「CPC 便宜」和「有没有成交」分开看，正是报告里最贵的那几个坑",
   "entry": "/internal/hit/funnel.html 投放日报判色器",
   "entry_status": "已接上线 2026-09-21"
  },
  {
   "algo": "industry_vitality",
   "demand": "D2 上海·中医馆",
   "output_sample": "行业生命力三维度 · 数据驱动/可替代/壁垒",
   "hookup": "/internal/geo/ 行业列 + /seo_pages/ 行业页结论块",
   "must_add": "三维评分目前无源数据，要么人工填，要么先接行业台账",
   "tracking": "同上",
   "value": "中高：76 个行业页每页一行结论，可被搜索引擎吃到",
   "entry": null,
   "entry_status": "未接（本批只上 4 个，按价值排序）"
  },
  {
   "algo": "three_d_profit",
   "demand": "D1 北京·KTV",
   "output_sample": "三维暴利指数 · 赛道值不值得进（乘积闸门）",
   "hookup": "/internal/geo/ 「赛道筛选」列",
   "must_add": "自然属性/复利/刚需三项要靠人工或行业台账赋值（无源数据）",
   "tracking": "同上",
   "value": "中高：abandon 布尔可直接做红灯",
   "entry": null,
   "entry_status": "未接（本批只上 4 个，按价值排序）"
  },
  {
   "algo": "five_dim_gold",
   "demand": "D1 北京·KTV",
   "output_sample": "五维金衡 · 长期价值分 + 逐维短板",
   "hookup": "/internal/geo/ 「项目筛选」列",
   "must_add": "同上，五维 0-50 需人工/台账赋值",
   "tracking": "同上",
   "value": "中：与三维暴利互补，做评分卡",
   "entry": null,
   "entry_status": "未接（本批只上 4 个，按价值排序）"
  },
  {
   "algo": "four_dim_landing",
   "demand": "D3 深圳·咖啡店",
   "output_sample": "四维度落地评估 · 资金/时间/能力/风险",
   "hookup": "落地可行性页 / 行业诊断页",
   "must_add": "四维 0-25 需人工填",
   "tracking": "同上",
   "value": "中：把「想做」翻译成「能不能落地」",
   "entry": null,
   "entry_status": "未接（本批只上 4 个，按价值排序）"
  },
  {
   "algo": "business_seven_layers",
   "demand": "D4 重庆·健身房",
   "output_sample": "商业7层体系 · 老板成熟度总表",
   "hookup": "诊断报告的收尾总表",
   "must_add": "七层 0-100 需人工填",
   "tracking": "同上",
   "value": "中：一次性总表，适合交付件",
   "entry": null,
   "entry_status": "未接（本批只上 4 个，按价值排序）"
  },
  {
   "algo": "risk_grade",
   "demand": "D4 重庆·健身房",
   "output_sample": "风险分级算法 · 偏离率×趋势 → 四级",
   "hookup": "★ 可立即上线：/internal/geo/ 表加一列「风险」，基准取同行业跨城市均值、趋势取 industry_dir（上行/下行/持平）",
   "must_add": "无需补数据——gap_top50.csv 里已有 score / industry_dir",
   "tracking": "同上",
   "value": "高：这是本批唯一「数据已经躺在硬盘上」的接法",
   "entry": "/internal/geo/ 「风险」列（每行一颗「算风险」）",
   "entry_status": "已接上线 2026-09-21"
  },
  {
   "algo": "nine_axis_locator",
   "demand": "D10 分身引擎推广路径",
   "output_sample": "九轴社会地图定位 · 雷达 + 坐标 + 画像",
   "hookup": "免费入口第一屏 / 分身引擎出口",
   "must_add": "九轴评分需问答采集；耗时 664ms 属可接受",
   "tracking": "同上；页面侧记录「进第一屏→完成九轴」转化",
   "value": "中高：免费入口的转化钩子",
   "entry": null,
   "entry_status": "未接（本批只上 4 个，按价值排序）"
  }
 ],
 "gray_list": [
  {
   "algo": "traffic_target",
   "demand": "D5/D6",
   "why": "priority 恒 P3、无淘汰词，平台词表是国内 8 平台 → 接不到「选词」下一步"
  },
  {
   "algo": "network_effect",
   "demand": "D8",
   "why": "density / distance_to_critical 口径未标单位 → 不能直接进决策"
  },
  {
   "algo": "pool_theory_eval / water_pool",
   "demand": "D3/D4",
   "why": "算法本身算得对，但判定阈值激进（满配才「有隐患」，没 🟢 档）→ 色块不能直接给用户看"
  },
  {
   "algo": "spread_validation",
   "demand": "D8",
   "why": "score/valid_rate 恒 1.0，只有 fraud_flag 有效 → 只能用半个"
  }
 ],
 "waste_list": [
  {
   "algo": "spread_validation",
   "demand": "D8",
   "why": "score 与 valid_rate 在任何输入下恒为 1.0",
   "verdict": "无",
   "evidence": "2026-09-21 八组输入实测：score 与 valid_rate 是同一个数（代码 valid_rate = score）；4 组（健康/中不溜/偏低）恒 1.0，只有「≥3 异常」的 3 组才掉到 0.0/0.2/0.8（另一组很低=0.8）。评分维度没有信息量",
   "value_from": "未知",
   "still_usable": "只剩 fraud_flag / anomalies 一个信号（3 档），要用只能当报警器，不能当分数"
  },
  {
   "algo": "overlay_matching",
   "demand": "D10",
   "why": "机会库只有 2 条常量，quadrant 恒「综合型」",
   "verdict": "无",
   "evidence": "机会库只有 2 条常量，quadrant 恒「综合型」（上一轮实测）",
   "value_from": "未知",
   "still_usable": "无"
  },
  {
   "algo": "spirit_engine",
   "demand": "—",
   "why": "注册即坏：未找到 BaseModule 子类（探针 / 与注册表都报）"
  },
  {
   "algo": "value_discovery_engine",
   "demand": "—",
   "why": "注册即坏：未找到 BaseModule 子类"
  }
 ],
 "updated_at": "2026-09-21 20:29:01 +0800",
 "entries_live": [
  {
   "algo": "risk_grade",
   "entry": "/internal/geo/",
   "where": "「缺口 Top 50」表新增「风险」列（每行一颗「算风险」按钮）",
   "map": "current=该行缺口分；baseline=同行业跨城市均值（本表 50 行口径）；trend=行业曲线（上行+0.5/走平0/下行-0.5）",
   "missing": "无（台账字段够用）",
   "shipped_at": "2026-09-21"
  },
  {
   "algo": "value_assessment_v123",
   "entry": "/internal/geo/",
   "where": "「缺口 Top 50」表新增「值不值得做」按钮（弹窗）",
   "map": "product_name=城市+行业；product_desc=该行城市/行业/曲线/需求/成本/分",
   "missing": "V1/V2/V3 的 12 个评分项（效率提升/省钱/覆盖规模/替代风险/保障感/愉悦感/彰显性/情绪持续/二手市场/共识强度/稀缺性/价值周期）台账里没有 → 弹窗默认 0.5（周期默认 1 年）并由人现场改，页面已写明",
   "shipped_at": "2026-09-21"
  },
  {
   "algo": "demand_verify",
   "entry": "/internal/hit/gate.html",
   "where": "立项闸门（6 问表单，人话问法）",
   "map": "6 问 → answers.q1_self_statement/q2_searched/q3_tried/q4_paid/q5_willing(0-100)/q6_action",
   "missing": "无（答案本来就是人填）",
   "shipped_at": "2026-09-21"
  },
  {
   "algo": "funnel_evaluate",
   "entry": "/internal/hit/funnel.html",
   "where": "投放日报判色器（12 指标 + 一行粘贴自动填）",
   "map": "data.platform/title + 12 个逐字指标名（CTR/展现量增速/完读率/停留时长占比/收藏率/评论率/点赞率/私信咨询率/加群率/关注率/转发率/裂变比）",
   "missing": "指标值来自日报本身（人工贴），算法不补数",
   "shipped_at": "2026-09-21"
  }
 ],
 "tracking": {
  "file": "/mnt/data/www/internal/hit/calls.jsonl",
  "written_by": "调度器 /opt/algorithms/algo_dispatcher.py 的 POST /algo/<id> 分支（备份 algo_dispatcher.py.bak_hitlog_20260921_2030）",
  "fields": [
   "ts 时间",
   "algo 算法",
   "src 来源页面(Referer)",
   "params 参数摘要(每层最多12键/字符串80字)",
   "ok 是否成功",
   "ms 耗时",
   "err 错误"
  ],
  "covers": "只记经 /algo53/algo/<id> 的 POST 调用；GET 查参数不计",
  "view": "/internal/hit/ 页面「埋点：今天谁在用」现场读该文件统计（今日调用数 / 今日成功失败 / 各算法 top / 近 7 天）",
  "why_this_way": "调度器原本把 log_message 关掉了（无访问日志），页面又都是静态页 → 调用计数只能在调度器这一层落盘；只加写日志，不动任何算法本体"
 },
 "broken_detail": [
  {
   "algo": "spirit_engine",
   "verdict": "坏 · 下架候选",
   "err": "未找到 BaseModule 子类",
   "evidence": "调度器 GET /algo/spirit_engine 返回 {\"ok\": false, \"error\": \"未找到 BaseModule 子类\"}；文件 /opt/algorithms/functional/spirit_engine.py 811 行，内容是分身后端逻辑（_get_conn/get_user_info/match_scenario/generate_response 等，走 pymysql）",
   "why_not_fixed": "它没有 run(data_bus) 这个算法契约；硬包一层只能用猜的输入输出，等于造假算法 —— 按纪律不接。要接先定义契约。",
   "action": "台账与 /internal/hit/ 页面标【坏 · 下架候选】；未改算法文件"
  },
  {
   "algo": "value_discovery_engine",
   "verdict": "坏 · 待修",
   "err": "未找到 BaseModule 子类",
   "evidence": "同样返回「未找到 BaseModule 子类」；文件 984 行，是 run_engine() 一次性生成 + save_to_db() 落库的脚本（calc_*/monte_carlo_stability）",
   "why_not_fixed": "没有入参契约，结果直接写库；改成可调用算法要先定输入输出。",
   "action": "台账与页面标【坏 · 待修】；未改算法文件"
  }
 ],
 "waste_verdicts": [
  {
   "algo": "spread_validation",
   "from": "未知",
   "verdict": "无",
   "evidence": "2026-09-21 八组输入实测：score 与 valid_rate 是同一个数（代码 valid_rate = score）；4 组（健康/中不溜/偏低）恒 1.0，只有「≥3 异常」的 3 组才掉到 0.0/0.2/0.8（另一组很低=0.8）。评分维度没有信息量",
   "still_usable": "只剩 fraud_flag / anomalies 一个信号（3 档），要用只能当报警器，不能当分数"
  },
  {
   "algo": "overlay_matching",
   "from": "未知",
   "verdict": "无",
   "evidence": "机会库只有 2 条常量，quadrant 恒「综合型」（上一轮实测）",
   "still_usable": "无"
  }
 ],
 "algorithms_broken_note": "两个都注册不上（未找到 BaseModule 子类）；原因与处置见 broken_detail。未修：它们不是算法模块，硬包就是假算法。",
 "waste_note": "废件 = 能跑但输出没有信息量。判定写在这里，证据挂在 waste_verdicts；/internal/hit/ 页面单列「不可用清单（别接）」。"
}