BATCH 95 — 上下文术语库／外部语域见证／搭配语义加固 — 2026 年 8 月 7 日

[PROMPT T 源语形态／语域见证模块]
Prompt T 必须先分析控制源，再选择目标术语。命名译者、词典、梵文对应、术语表或公共译本只是可能语域／义项的见证，不能覆盖局部原文语法。

BATCH95 箭头语义优先规则：继承规范表中的“源词项 → 目标译法”只是有条件速记——只有本处先被判定为相应技术义项、符合选定语域，且局部句法支持时，才采用该保护译法。箭头绝不表示“所有出现都替换成同一字符串”。若有表面冲突，以 Batch95 的源文角色、形态、搭配与直接源文分析为准；有依据的上下文偏离必须记录，不得作为未说明的风格偏好。

每个高风险藏文 lemma／词簇在选译前记录：source-unit ID 与精确藏文；lemma 与切分；属格、施事／工具／因由、处所、名词化、并列与分句边界；复合边界与标题／专名状态；名词、动词、形容词、关系、语法化、同位、零系词或未决语法角色；技术教义、普通词义、标题成分、技术配对／复合、空间／比喻／描写、意旨／旨趣／意义或异常等语义／篇章角色；直接见证；外部语域见证；实际给出的梵文对应；实际给出的英文 gloss；英文栏究竟是明确、空白、星号、仅音译或缺失；局部搭配测试；最终目标译法与置信度。

绝不能把藏→梵对应自动写成外部译者的英文 gloss；星号／空白英文栏也不是英文译词；同一藏文 lemma 不得全局强制一个英文 gloss。

回归见证：
- dbyings：Wisdom/Malcolm Lesson 6 表中 Tibetan=dbyings、Sanskrit=dhātu、English=*；证明梵文对应，不提供英文 gloss。技术、dharmadhātu 与空间义逐例判断。
- dgongs pa：Wisdom 大圆满术语材料明确在“不间断观修状态”义项使用 transcendent state；这是义项见证，不是普遍等号。标题及意旨／旨趣义须局部分析。
- sems kyi rig pa / rig pas na：先判名词关系、格功能与分句结构，再决定英文是否动词化；不得为自然化而补造独立的 mind/knower。

搭配失败规则：若首选词表术语与其中心词、依附短语、标题、跋文或谓词组合后语义异常，必须先重新分类该处的义项／角色，而不是继续“润色”。

MODULE_ID = B95-TIBETAN-REGISTER-WITNESS-1.1
PROMPT_T_RUNTIME_SHA256 = [从此精确文件计算并在报告中记录]


BATCH 94 — 藏文语域锁定／精确复合词／形态分析／安全替换／跨提示词同步加固 — 2026 年 8 月 6 日

范围与继承

本补丁适用于藏文／印度语系翻译、比较审校、润色、出版、Blogger 排版、严格 HTML QA、控制器、打包及无修复晋升工作流。它继承 Batch93 及此前全部控制源、见证分类、隐藏主体、谓词拆分、直接见证、Prompt 6/9 执行证据、真正累积日志、规范包、渲染完整性与同轮状态纪律。

角色边界

Prompt T 是唯一规范的藏文语域模块。Prompt 1 建立译前逐例矩阵；Prompt 6 以原文核验；Prompt 9 对每个修复句作自然度与无回退通读；Prompt A 与 Universal Intro 保存 source-unit ID 并把可见文本修复转交 Prompt 1/6/9；控制器验证 Prompt T 的存在、版本与哈希；排版器只盘点风险，不得作教义术语替换；严格 QA 检查全部交付物，并对陈旧政策或失效夹具判 FAIL。

PROMPT_T_REQUIRED = Prompt_T_Tibetan_Indic_Source-Anchor_Verification_Public-Translation_Contamination_Repair_v1_17.txt
PROMPT_T_MODULE_ID = B95-TIBETAN-REGISTER-WITNESS-1.1
PROMPT_T_RUNTIME_SHA256 = [calculate from this exact file and record in reports]

A. 规范模块／语域声明
- 声明 SOURCE_LANGUAGE、SOURCE_REGISTER、TARGET_LANGUAGE、TARGET_REGISTER、CONTROLLING_SOURCE、SELECTED_TRANSLATOR_REGISTER、EDITABLE_WORKING_BASE、SECONDARY_WITNESSES、PROJECT_TERMBANK 与 PROMPT_T_SHA256。
- 只有当用户选用 Ācārya Malcolm Smith 术语且 TARGET_LANGUAGE=English 时，才把该表作为受保护的“藏文→英文”语域表；其他目标语言只能把它当作语域见证，不得照搬英文目标词。
- 命名译者语域只是术语／文体权威，绝不能取代控制源和局部句法。
- Prompt T 缺失、不可读、版本不符或哈希不符时，必须明确 FAIL 或声明后备方案；不得声称已执行。

B. 逐例精确矩阵
每一相关处记录：source-unit ID、精确藏文、切分、助词、语法角色、选定译法、拒绝译法、命名语域证据、局部理由、置信度、成品出现次数、陈旧形式次数。Prompt 1 建立，Prompt 6 核验，Prompt 9 通读。

C. 词元／助词／复合边界
切分 དང་、属格 གི་/ཀྱི་/གྱི་/འི་/ཡི་、施事／工具格、否定词、助动词与分句边界。把相邻表达分类为并列、属格、同位、词汇复合、层叠修饰、谓词序列或未决簇。不得仅因藏文相邻就添加英文 “and”“of”、等同关系或破折号同位。未决簇至少保留两种解析、证据、置信度与出版处理。

D. 词汇复合与并列谓词
真正并列的 རིག་ཅིང་གསལ་ 要保留两个谓词；词汇簇 ཤེས་རིག་ 不得机械拆开。Batch93 谓词拆分不得反向机械套用，须结合局部语法与选定语域证据。

E. 光明／明晰区分及有证据例外
选定英文语域下：འོད་གསལ་ ('od gsal) → luminosity；独立 གསལ་བ་ (gsal ba) → clarity/clear；རང་གསལ་ (rang gsal) → self-clarity/self-clear。不得一律扁平化。若直接局部文献证据表明 gsal ba 是 'od gsal/prabhāsvara 的省略，则可用 luminosity，但必须记录证据与例外，不得盲目套表。

F. rang 词族形态
逐例分析 rang byung、rang shar、rang gsal、rang grol、rang rig 与层叠 rang 复合。不得把 self-originated 改成 self-arisen、把 self-clarity 变成悬空形容词、把 self-liberation 变成自主施事者，或把反身／局部表达实体化为形而上 Self。

G. 禁止全局教义替换
未先盘点每一处，不得对受保护藏文／印度语系术语族作全局或正则替换。脚本修改后，逐句回到控制源，记录搜索词、替换词、次数、语法角色、邻近复合词、意外变化、陈旧形式及 Prompt 9 通读结果。

H. 从原文局部重析
旧译若提供藏文并未明示的句法分组，必须重析整个藏文分句，不得只在旧句法中换术语。强制夹具包括魔／鬼障句、本尊／悉地句、认知与显现、བྱེད་མི་ཤེས་ 情态、对治法施事，以及池塘／船只见证控制。

I. 受保护语域的自然度
保留受保护术语，同时修复难读名词堆、虚构领属、悬空结构、无先行词、冠词制造的实体及伪同位。不得暗中替换受保护术语。机械修改的每句都须全文重读。

J. 跨成品陈旧政策扫描
扫描正文、注释、Prompt 6 报告、Prompt 9 报告、术语账本、累积日志、README、manifest、提示词包与 ZIP 成员中的过时形式及互相矛盾的现行政策。

K. 正向预期命中／夹具有效性
夹具既要验证拒绝形式不存在，也要在控制源确有该表达时验证当前所需形式存在。零命中不自动 PASS。记录源单元、当前预期、拒绝形式、实际命中及夹具是否命中精确成品。过时或锚点错误的夹具判 FAIL。

L. 依赖与排版器边界
Prompt T 不可用时控制器须 FAIL 或明确后备。排版器只能盘点并保留术语；可见术语修复必须经 Prompt 1/6/9 并写入累积日志。

M. 包同步
Prompt T 文件名、版本、模块 ID 与 SHA-256 必须在控制器、报告、独立 TXT、公共页嵌入副本、Blogger 片段、README、manifest 与 ZIP 成员之间一致。禁止互相漂移的复制术语表。

N. 同轮状态
任何术语、解析、依赖、报告、渲染、manifest、哈希、ZIP 或打包修复均属实质修复；同轮不得 FINAL。

BATCH94 强制确认行
Batch94 藏文语域锁定／精确复合词／形态／安全替换／跨提示词同步门槛已完成：已声明规范 Prompt T 与命名译者语域；逐例应用精确映射；区分 rig pa、ye shes、shes pa 与 shes rig；区分 'od gsal、gsal ba 与 rang gsal，并记录有证据的局部例外；按语境保护 thig le 与 thig le nyag gcig 为 tilaka 与 the sole unique tilaka；rang 词族未因全局替换而漂移；相邻藏文未自动变成 “and”“of”、等同或同位；旧译句法已从藏文重析；机械修改句已在 Prompt 9 下重读；陈旧政策与正向命中夹具已核验；提示词依赖已验证；排版器边界已保留；同轮修复不得晋升。


PROMPT T v1.16 — 规范“藏文→英文”语域模块
MODULE_ID = B95-TIBETAN-REGISTER-WITNESS-1.1

启用与权威
- 仅当 TARGET_LANGUAGE=English 且 SELECTED_TRANSLATOR_REGISTER 明确指定 Ācārya Malcolm Smith 或由其派生的本项目英文语域时，才启用下列受保护英文映射。
- 本模块服从控制藏文与局部句法，不授权字典式替换、见证污染或静默改原文。
- 非英文目标只能把本模块作为语域见证，并另建目标语言矩阵。

基础术语
- རིག་པ་ rig pa → vidyā；首次可解释为 “knowledge (vidyā)”。拒绝一律 awareness 或 consciousness。
- མ་རིག་པ་ ma rig pa → ignorance。
- ཡེ་ཤེས་ ye shes → pristine consciousness；与 rig pa、rnam shes / shes pa 区分。
- ཤེས་པ་ shes pa → 名词时 consciousness；动词／事件时 know/knowing，依句法。
- ཤེས་རིག་ shes rig → 作为已见证词汇簇时用 knowing consciousness；不得机械拆成 “consciousness and cognizance”。
- འོད་གསལ་ 'od gsal → luminosity。
- གསལ་བ་ gsal ba → 独立使用时依语法 clarity/clear；若直接局部证据表明是 'od gsal/prabhāsvara 省略，则用 luminosity 并记录证据。
- རང་གསལ་ rang gsal → self-clear/self-clarity。
- ལྷུན་གྲུབ་ lhun grub → natural perfection/naturally perfect；除非用户另选语域或原文专门要求，不得漂移为 spontaneous presence。
- ཐིག་ལེ་ thig le → 所选大圆满语域中用 tilaka；若控制语境明确是身体 bindu/drop、精质、书写点等，不得强制 tilaka，须记录例外。
- ཐིག་ལེ་ཉག་གཅིག་ thig le nyag gcig → the sole unique tilaka。
- རང་བྱུང་ rang byung → self-originated。
- རང་ཤར་ rang shar → self-arisen。
- རང་གྲོལ་ rang grol → 依句法 self-liberation/self-liberated。
- རང་རིག་ rang rig → 局部解析；不得自动制造形而上 self-knower。仅在句法及语域证据支持时，才可用 personally intuited vidyā / personally intuited pristine consciousness。

受保护复合／词簇
- རང་བྱུང་ཡེ་ཤེས་ → self-originated pristine consciousness。
- རང་བྱུང་རང་གསལ་ → self-originated self-clarity。
- རང་བྱུང་རང་གྲོལ་ → self-originated self-liberation。
- རང་བྱུང་རང་ཤར་རང་གསལ་ → 分别保留 self-originated、self-arisen、self-clear。
- རིག་པ་ཡེ་ཤེས་ → 未决紧缩词簇，除非局部证据解决关系；不得自动加 “and”、等同、破折号同位或属格。无 དང་／明示属格时，任何关系词须标注为暂定。
- རང་རིག་ཡེ་ཤེས་ཐིག་ལེ་ཉག་གཅིག་ → 局部切分并说明，不得虚构 “of”。审校文本中 “pristine consciousness, the sole unique tilaka” 比 “the sole unique tilaka of pristine consciousness” 更安全，但只是局部先例，不是冻结整句。

其他项目术语
- བསྙེན་སྒྲུབ་ → approach and accomplishment；区别 བསྐྱེད་རྫོགས་ generation and completion。
- གཟུང་འཛིན་ → 明确成对时 apprehended and apprehender。
- དགེ་སྡིག་ → virtue and nonvirtue。
- བདེ་གཤེགས་སྙིང་པོ་ → 本项目语域 sugata essence；sugatagarbha 若只是惯例回译须披露。
- སངས་རྒྱས་སྙིང་པོ་ → 局部句法支持时 buddha-nature。
- བྱིང་རྨུགས་འཐིབ་རྒོད་ → 保留全部项目；审校译法 laxity, drowsiness, mental fog, and agitation。

强制形态测试
rig pa ye shes；rang rig ye shes thig le nyag gcig；shes rig；gsal rig；rang byung ye shes；rang byung rang gsal；rang byung rang grol；rang byung rang shar rang gsal。

强制源句法夹具
- 魔／鬼障句：重新解析藏文，不继承旧英文分组。
- 本尊／悉地句：重新解析藏文，不继承 “favorable appearance”等译文生成关系。
- དུ་མ་རིག་ཅིང་གསལ་：保留多样认知／能知与明晰，不得译为 “manifests as everything”。
- བྱེད་མི་ཤེས་：保留 “does not know how”，不得强化为客观不可能。
- 对治法施事：保留对治法能／不能克服什么，不得倒置施事。
- 池塘乌鸦见证：控制源为 pond 时，不得导入熟悉的 ship 读法。

正向夹具规则
每个夹具记录源单元、精确原文、当前预期、拒绝形式、正／负命中及精确成品匹配。除非证明源表达确实不存在，否则零命中不算 PASS。


BATCH 93 — 跨语言“知的功能”／隐藏主体／谓词拆分／提示词执行证据／渲染夹具加固 — 2026 年 8 月 2 日

范围与继承

本补丁适用于全部翻译、比较审校、原典复原、全书润色、术语库、出版、Blogger 排版、打包与严格 QA 工作流，也适用于所有源语言—目标语言组合，包括藏文、梵文、中文、英文、德文、法文、日文及多语比较项目。

Batch93 继承且不削弱此前关于控制源权威、见证材料分类、防止见证污染、遗漏／增添／极性／范围／情态／施事／领属／反身审查、术语族一致性、全书润色、累积修复保存、精确成品回读、状态纪律、同轮不得晋升 FINAL、Blogger 输出模式分离、规范包身份、渲染证据及审计完整性的全部要求。

角色边界：各门槛必须按提示词实际职责执行。翻译与审校提示词负责源文—译文分析和修复。Blogger 排版器仍是排版工具：它只盘点高风险术语，并把可见教义或文字改动转交 Prompt 6 与 Prompt 9，不得暗中充当翻译提示词。严格 HTML QA 可以发现并记录术语／文字风险，但涉及可见文本的实质修复，在适用时必须通过 Prompt 6 与 Prompt 9。

目的：防止源文中的过程、功能、性质、状态、显现、关系或认知，仅因目标语言的冠词、名词化、代词或文体习惯而变成隐藏的施事者、观察者、能知者、拥有者、见证者或形而上实体。

A. 源语言—目标语言声明门槛

每项翻译或审校任务必须明确声明：
SOURCE_LANGUAGE = […]
SOURCE_REGISTER = […]
TARGET_LANGUAGE = […]
TARGET_REGISTER = […]
CONTROLLING_SOURCE = […]
EDITABLE_WORKING_BASE = […]
SECONDARY_WITNESSES = […]
PROJECT_TERMBANK = […]

语体须在适用时具体说明，例如古典藏文、现代藏文、文言文、现代简体中文、佛教混合梵文、现代／文学英文、学术德文、现代日文等。不得假设同一语言的历史语体或文体可以互换。

必需证据：
- 已声明源语言：PASS/FAIL。
- 已声明源语体：PASS/FAIL。
- 已声明目标语言：PASS/FAIL。
- 已声明目标语体：PASS/FAIL。
- 已识别控制源：PASS/FAIL。
- 已分类次要见证：PASS/FAIL。
- 提示词、报告、文件名与 manifest 的语言身份一致：PASS/FAIL。

B. “知的功能”与“能知者”门槛

对每一处涉及知、认知、辨认、觉知、感知、理解或辨别的高风险表达，先把语法与语义角色分类为：
1. 知的动作／事件；
2. 认知过程；
3. 能知功能／能力；
4. 明知、清明或能认知的性质；
5. 觉知／辨认状态；
6. 教义技术术语；
7. 源文明确提出的能知者／认知主体；
8. 源文特意引入、以便寻找、分析、否定或解构的主体。

不得把过程或性质自动变成“能知者”“知道的人”“认知主体”“内在观察者”“见证者”“拥有经验的觉知”“作为独立实体的意识”、Self 或背景体验者。危险形式包括中文的能知、能知者、能知之识、所知者、觉知者；英文 the knower、the knowing one、the awareness、an observer、that which knows；德文 der Wissende、被实体化的 das Gewahrsein、das erkennende Subjekt；法文 le connaissant、celui qui sait、被当作实体的 la conscience；日文 知る者、認識主体、覚知者；以及其他施事名词、定指名词短语、名词化分词或人格化结构。

这些形式并非一律禁止。逐例判断其是否由控制源明确要求、语法上必要但语义中性、属于允许的自然化，或是无根据的隐藏主体。源文若只表达功能／事件／性质，译文须保持该角色；源文若明确提出见者、闻者、知者、作者、禅修者或观察者，以便后文解构，则必须保留该修辞主体。

必需证据：
- 检出的知／认知术语族：[…]。
- 各处语法角色：[…]。
- 源文明确主体：[…]。
- 译文新增施事名词：[…]。
- 尚存无根据隐藏主体：[数量／清单]。
- 已保留源文明示的解构性主体：PASS/FAIL。
- 知的功能／能知者门槛：PASS/FAIL。

C. 过程／功能／性质／实体分类门槛

对每个教义敏感表达，判断它在源文中是过程、事件、活动、功能、能力、性质、状态、关系、显现、认知、认知对象／内容、施事者、实体，或必须保留惯用名词形式的技术术语。译文不得悄然改变类别。

高风险转移包括：知→能知者；见→见者；体验→体验者；明晰→发光实体；觉知→独立见证者；显现→某个“显现者”；自知→一个知道自己的 Self；空性→绝对基体；不二→一个形而上实体；关系→独立存在的属性；作用→自治行动者；认知→永久容器式意识。

必需证据：
- 检出的高风险语法类别转移：[…]。
- 每处源文角色已识别：PASS/FAIL。
- 译文角色已保留：PASS/FAIL。
- 无根据实体化：[数量／清单]。
- 过程／功能／性质／实体门槛：PASS/FAIL。

D. 并列谓词拆分门槛

源文若并列多个不同谓词、性质、功能、对象或教义类别，须保留其正确数量与关系。不得把两个谓词合成一个方便的目标语复合词、重复翻译一个谓词、遗漏并列项、把一个谓词变成其他谓词的主语／拥有者、把三项强套成无关的“三法”、因目标语有优雅总称便推断同义，或无理由拆成互不相干的句子。

例：能知而明、空而鲜明、显而不成立、认知／明晰／空、具有或不具有认知、见闻知、起住灭。藏文 རིག་ཅིང་གསལ་ 分别表达知／认知与明晰；中文“明知”或英文 luminous awareness 可能把两者压成一个概念，必须按语境审查。

必需证据：
- 源文预计并列项：[数量／清单]。
- 译文保留并列项：[数量／清单]。
- 复合词压缩风险：[数量／清单]。
- 重复翻译风险：[数量／清单]。
- 主语制造风险：[数量／清单]。
- 谓词拆分门槛：PASS/FAIL。

E. 认知／感知／显现／存在门槛

盘点并区分：知、认知、辨认、觉知、意识、感知、见、显相、显现、发生、展现、表现能量、存在、临在、明晰、光明、自反／自知。不得仅因某个目标语灵性术语听起来高雅，就把一个语义族替换成另一个。

禁止无根据转移：认知→显现；显现→认知；知多种→显现为一切；明晰→觉知；觉知→意识；意识→永久见证者；辨认→全知；存在→大写 Presence；显相→独立存在对象；自知→Self 知道自己。源文若同时含认知与显现语言，两者都要保留。

必需证据：
- 认知术语族：[…]。
- 感知术语族：[…]。
- 显现术语族：[…]。
- 存在／临在形式：[…]。
- 无根据的语义族替换：[数量／清单]。
- 认知／感知／显现／存在门槛：PASS/FAIL。

F. 量词与分配性门槛

分类一、二、多、种种、若干、各、每、一切、全部、无、某些、部分、整体、普遍、集合、分配等形式。不得把“多／种种”译成“一切”，把“某些”译成“全部”，把“各”抹成集合整体，把复数集合实体化为一个统一实体，把一般陈述夸大为全知，或把明确普遍命题削弱为几个例子。除非源文支持，不得增添 all、always、entirely、completely、everywhere、never、nothing whatsoever 等全称词。

必需证据：
- 源文量词：[…]。
- 译文量词：[…]。
- 无根据普遍化：[数量／清单]。
- 源文明示全称被削弱／遗漏：[数量／清单]。
- 量词／分配性门槛：PASS/FAIL。

G. 认识论力度／情态／可能性门槛

区分：不知道如何、不理解、不能、缺乏能力、不可能、不发生、不生、不成立、不存在、尚未、禁止、不必要、无效、不确定。不得把“不知如何”加强为“不能／不可能”，把不可能削弱为不确定，把缺席变成禁止，把不成立变成不显现，把“尚未”变成绝对否定，或把认识论限制变成存在论断言。

必需证据：
- 检出的情态／认识论结构：[…]。
- 源文力度分类：[…]。
- 译文力度比较：PASS/FAIL。
- 被加强的情态：[数量／清单]。
- 被削弱的情态：[数量／清单]。
- 认识论力度门槛：PASS/FAIL。

H. 领属／反身／Self 语言门槛

审查 its own、one’s own、itself、oneself、self-knowing、self-awareness、self-luminous、self-manifesting、inner、inherent、intrinsic、belonging to、possessed by 及相应目标语形式。不得因目标语偏好领属句法而添加拥有者；不得把每个反身形式都理解为形而上的 Self、自我作用实体、拥有经验的内在觉知或主客式反身认知。

判断源文究竟表示：自然而然／自行、在其本处、从自身、直接／当下、自生、反身认知、普通意义的“自己的”，或专门教义反身术语。

必需证据：
- 源文反身／领属形式：[…]。
- 译文新增拥有者：[…]。
- 无根据的 Self 式读法：[数量／清单]。
- 反身性在不实体化下得到保留：PASS/FAIL。
- 领属／反身门槛：PASS/FAIL。

I. 不扭曲源文的目标语自然度

Prompt 9 全书润色必须检查反实体化修复是否制造了晦涩或不自然的译文：残句、抽象名词堆积、机械的“知”、不自然被动语态、不必要删除主语、刻意回避普通代词、伪古典文体，或普通语言更准确时却堆砌教义术语。每次修复都须平衡源文忠实、语法自然、教义非实体化、术语一致与声明语体中的可读性。非施事表达若变得模糊或改变修辞力度，并不自动优于自然表达。

必需证据：
- 反实体化修复已在完整语境重读：PASS/FAIL。
- 新增翻译腔：[数量／清单]。
- 被不当删除的必要主语：[数量／清单]。
- 目标语自然度门槛：PASS/FAIL。

J. 直接见证与文本异常证据层级

遇到异常拼写、疑似讹误、语法不规则、缺文、分段不明、版本冲突、OCR 损坏、编辑规范化或不明标点时，依次使用：
1. 已提供的控制源；
2. 同一文本经核实的直接见证；
3. 经核实的异本／版本；
4. 同一见证的可靠精确转录；
5. 明确讨论该段的源语言注释；
6. 同一语料中的近似平行段；
7. 字典与语法书；
8. 其他语言译本。

译本只能作为次要见证，不能暗中覆盖原文。不得静默改订。每项推定读法必须记录问题原形、建议读法、证据、见证类别、置信度、可见译文是否改变、是否需要方括号／注释。没有确定文本关系时，不得把平行段称作“手稿异文”。

必需证据：
- 检出的文本异常：[…]。
- 查阅的直接见证：[…]。
- 采用的推定改订：[…]。
- 无根据静默改订：[数量／清单]。
- 文本异常证据层级：PASS/FAIL。

K. 语言特定应用说明

通用门槛必须按真实源语言与目标语言应用，不得机械移植中文专用禁则。

藏文→中文：检查动词性 རིག་／ཤེས་ 是否变成能知者／能知之识；并列时区分 རིག་ 与 གསལ་；认知词汇本身不得推导出“显现为一切”；避免未限定“无知”等日常义与技术语境冲突的词。

藏文→英文：检查冠词是否制造 the awareness／knower／observer；检查 it 是否引入藏文没有的实体；不得用 awareness 一词笼统覆盖多个藏文词；源文是动词性知时尽量保持动词性。

中文→英文：不得把知、觉、觉知、灵觉、明觉、觉性全部压成 awareness；不得假设 者 永远表示永久主体；检查英文冠词与大写是否制造形而上主语；源文论证依赖明确人物角色时要保留。

英文→中文：先分类 awareness、knowing、consciousness、observer、witness，再选择中文；不得自动把 knowing 译成能知，或把 knower 固定为永恒能知者；适当把英文名词化还原为自然中文动词；保留真实主语。

德文／法文／日文及其他语言：识别该语言自己的施事名词、冠词、语法性别、代词与名词化模式，并记录可能暗示独立认知主体的目标语形式。例子只说明通用门槛，不取代语境分析。

L. Prompt 6／Prompt 9 执行证据门槛

只说“使用了 Prompt 6 和 Prompt 9”不够。每项实质翻译审校必须提供两个可分别识别的证据部分。

PROMPT 6 — 源文忠实报告：记录精确控制源、可编辑工作底稿、已分类见证、分段方法、遗漏、增添、极性、施事、领属、反身、范围／量词、情态、并列谓词、技术术语族、文本异常、未决不确定性及修复。

PROMPT 9 — 全书润色报告：记录不间断全书重读、目标语自然度、重复术语一致性、代词与先行词、施事／隐藏主语一致性、名词化／翻译腔、节奏与重复、标题与结构标签、书名／叠句／跋文一致性、读者歧义及出版风险。

还须记录 Prompt 6 与 Prompt 9 的准确版本和文件 SHA-256、源／目标语言与语体，以及相关门槛的 PASS／FAIL／NOT APPLICABLE。最终 QA 必须指向这两个证据部分。

必需证据：
- Prompt 6 执行有证据：PASS/FAIL。
- Prompt 9 执行有证据：PASS/FAIL。
- 提示词版本与哈希已记录：PASS/FAIL。
- 两份报告中的语言组合正确：PASS/FAIL。
- 提示词执行证据门槛：PASS/FAIL。

M. 真正累积日志门槛

称为“累积变更日志／累积源锚账本”的文件必须保留此前条目、追加本轮变化、标记被取代决定、解释反转、在重要处记录旧措辞与新措辞、正确识别基线轮次，并确保没有互相矛盾的术语政策同时保持有效。仅含本轮差异的文件必须称“本轮变更日志／本轮修复日志”，不得称“累积”。

必需证据：
- 既往累积历史已保留：PASS/FAIL。
- 最新差异已追加：PASS/FAIL。
- 被取代决定已标记：PASS/FAIL。
- 相互矛盾的有效政策：[数量／清单]。
- 累积日志门槛：PASS/FAIL。

N. Blogger 背景层决策门槛

改变背景前，必须区分：(1) 文章外的网站／主题背景；(2) Blogger 自有页面／文章／内容容器；(3) 文章 fragment；(4) 文章内部卡片、面板与引文区。记录每层期望状态。

不得为了白色阅读画布而移除站点主题；只需改文章卡时却漂白广泛 Blogger 祖先元素；未检查实时 DOM 就对宿主使用 background-image:none；让一个 article fragment 覆盖其他文章／组件；或把内部 feature panel 与宿主文章背景混为一谈。需要白色阅读画布时，锁定最窄必要宿主，除非用户要求否则保留外部主题及原生标题／日期，核验桌面与手机 computed background，并记录所有被选择的 Blogger 自有 selector。

必需证据：
- 背景层已映射：PASS/FAIL。
- 各层期望背景：[…]。
- 定向 Blogger selector：[…]。
- 是否存在广泛宿主主题覆盖：YES/NO。
- 桌面 computed background 已检查：PASS/FAIL。
- 手机 computed background 已检查：PASS/FAIL。
- Blogger 背景层门槛：PASS/FAIL。

O. 渲染夹具完整性门槛

远程图片、字体、样式表、脚本或其他资产离线无法加载时，允许：下载并使用完全相同的资产字节；使用 URL 与哈希均匹配的已验证缓存；在缺失资产下渲染并披露限制；提供静态 DOM／CSS 证据；要求在实时发布环境最终确认。

禁止：把旧的整页截图当作替代背景；把含页面文字的截图当 CSS 夹具；使用仅仅相似的图片却声称精确主题渲染；暗中复用陈旧截图；只为生成证据而修改交付 HTML 却不记录差异；或在仅使用静态 renderer 时声称完成实时浏览器检查。

每个夹具须记录请求 URL、夹具文件名、MIME 类型、来源、SHA-256、验证方法、是否为精确资产、交付 HTML 是否改变、渲染时间及被渲染 standalone 的准确哈希。

必需证据：
- 使用的渲染夹具：[…]。
- 精确资产验证：PASS/FAIL/NOT APPLICABLE。
- 是否把页面截图当资产：YES/NO。
- 陈旧渲染证据：[数量／清单]。
- 被渲染 standalone 的准确哈希：[…]。
- 渲染夹具完整性门槛：PASS/FAIL。

P. 同轮晋升纪律

可见翻译、术语、源文解释、隐藏主体、谓词拆分、自然度、源注、Prompt 6／9 证据、累积日志、Blogger 背景、渲染夹具、报告、manifest、哈希或 ZIP 修复均属实质修复。任何一项发生，本轮状态必须保持 REPAIR。只有后来一次精确成品审查重新打开完整前一包，并对可见文字、CSS、DOM、证据、元数据、包内容和哈希都不作任何改变，才可晋升。

Batch93 强制确认行：
Batch93 跨语言知的功能／隐藏主体／谓词拆分／提示词执行／渲染夹具门槛已完成：已声明源语言、目标语言及语体；已区分知的功能与能知实体；已分类过程、功能、性质、状态、关系、施事与实体角色；并列源文谓词未被目标语复合词压缩；已区分认知、感知、显现、显相、存在、临在、明晰与意识术语族；量词范围已保留；认识论上的“不知如何”未被加强为客观不可能；领属与反身已审查隐藏 Self／拥有者插入；源文明示的解构性主体已保留；文本异常依直接见证层级处理；Prompt 6 与 Prompt 9 执行分别有证据；累积日志已验证为真正累积；Blogger 背景层已分类；渲染夹具已验证为精确资产而非页面截图；同轮修复晋升已禁止。

BATCH 90 — 覺 术语族上下文加固：覺性 / 覺者 / 靈覺 / 覺知 区分 — 2026年7月7日

适用范围与继承：
Batch90 继承 Batch89 覺性 / 圓覺性 语域加固以及 Batch76–89 全部规则。本补丁加入更细的 覺-family 上下文区分。

核心教训：
不得把所有 覺、覺性、覺者、靈覺、覺知 等出现机械视为同一个语义单位。同一个字根可能承担不同角色：
- 覺性：根据语境可为 awakened nature / awakened awareness / nature of awakening。
- 覺者：当人物/主体角色明确时，是觉者、觉醒者或 Buddha 角色。不能只压缩成 “awakened”。
- 靈覺：描述活生生的觉知功能时，可考虑 numinous awareness / luminous awareness / numinous knowing；除非真的是觉醒事件，否则避免 “miraculously awakened”。
- 覺知：按语法角色可译 awareness、knowing、cognizing 或 being aware；不可自动译成 awakening。

[覺者 / 靈覺 / 覺知 上下文关卡]
Bloodstream 类文本示例：
「此土云覺性。覺者靈覺」
这里是在说明佛是自心的觉知、响应、灵妙觉用：回应、感知、挑眉、眨眼、动手动脚等。此处若译成“觉者是奇妙地觉醒”，会丢失功能性觉知义。应保留 awareness-function，例如 “the awakened one is numinous awareness” 或根据句法译为“觉者即灵觉之作用”。

[避免把功能性觉知误读为觉醒事件]
当 覺 修饰 覺者 等人物角色词时，必须判断是“觉醒者/觉者”还是在解释觉知功能、本性作用。不可把 覺性 的默认译法机械套到所有 覺 复合词，也不可把功能性 awareness 变成突然发生的 awakening event。

[完整 覺-family 扫描]
发现一个 覺-family 问题时，扫描：覺、覺性、覺者、覺悟、覺知、靈覺、明覺、妙覺、本覺、始覺、圓覺、佛覺、覺察，以及相关复合词。记录源文角色、语法角色、选择译法、否定替代译法与理由。

强制 QA/changelog 行：
Batch90 覺-family 上下文关卡已检查：已区分 覺性 / 覺者 / 靈覺 / 覺知 角色；已检查 awareness-function 与 awakening-event 区别；人物角色译法未被削平；相关处已执行整个术语族扫描。

BATCH 89 — 中文「覺性 / 圓覺性」awakened-awareness 语域加固 — 2026年7月7日

适用范围与继承：
本 Batch89 加固补丁适用于所有 Awakening to Reality / John Tan / Soh Wei Yu 翻译、审校、精修、对抗性审计、比较审计、累积日志、源文恢复、无修复晋级、prompt-manager、打包与交接工作流。它继承且不削弱 Batch76–88，包括 Self/self 语义大小写、术语库执行、source-role 分类、选择性合并、累积修复保留、视觉 QA、状态纪律、无修复晋级、术语族一致性、中文 體-family 语域、受 的上下文译法，以及防削薄源文关卡。

来自「覺性」语域审查的核心教训：
覺性 不能用一个固定英文词机械处理。安全的英文范围必须按源文角色和教义语境判断：当它指明觉的明知能力/觉醒能力本身时，可用 “awakened awareness” / “awakened knowing”；当 性 的义值必须显出时，用 “nature of awakening” 或 “nature of enlightenment”；在 圓覺經 / Perfect Enlightenment 语境，如 圓覺性、淨圓覺性、隨順覺性 等，则用 “nature of perfect enlightenment”。外部佛教译本可帮助判断英文语域，但上传的中文源文仍是控制权威。

[中文「覺性 / 圓覺性」语域关卡]
对于中文佛教源词 覺性、圓覺性、淨圓覺性、隨順覺性、覺、圓覺、淨覺、覺知、本覺、始覺，以及邻近的觉知/觉醒复合词，不得用一个英文等值词全局机械替换。必须逐处判断：
- 当 覺性 指明觉的明知 / 觉醒能力本身时，可用 “awakened awareness” 或 “awakened knowing”。
- 当 性 具有教义作用、必须保留 “nature” 义值时，用 “nature of awakening” 或 “nature of enlightenment”。
- 当语境明确属于 圓覺經 / Perfect Enlightenment，尤其出现 圓覺、淨圓覺性、圓覺性、隨順覺性 等表达时，用 “nature of perfect enlightenment”。
- 只有在 佛性 / tathāgatagarbha 明确出现，或本地源文/注释清楚把 覺性 等同于 佛性 时，才可用 “buddha-nature”。不得把 “buddha-nature” 作为 覺性 的默认译法。
- 只有在实践/禅修语境中，且 性 的力度较轻时，才可用普通 “awareness”；“mindful awareness” 只适用于源文确实讨论正念/注意训练的情境，不可泛用于觉性。
- 除非源文明确是在谈 心識、唯识/阿毗达磨的心识分析，或类似 非覺性 这类需要 “consciousness” 的教义语境，否则避免用 “consciousness”。不得把 覺性 译成 “the Self”、“soul”、“pure consciousness”、“eternal witness”，或任何听起来像永恒形上见知者的表达。

[覺性 外部语域参考警戒]
BDK / Peter Gregory 与 Charles Muller / Perfect Enlightenment 风格的译本，可帮助判断 圓覺經 相关短语的英文语域，如 “nature of Perfect Enlightenment” 或 “nature of enlightenment”。学术用法 “awakened awareness (juexing 覺性)” 可支持 “awakened awareness” 作为可用默认项。84000 对 rig pa / vidyā 与 ye shes / jñāna 的术语处理，只能在藏文/梵文相邻源文角色中提供参考；它不是中文 覺性 的直接权威。除非 source-role 证据明确要求，不得默认把 覺性 等同于 rigpa/vidyā。外部来源只是语域见证，不得覆盖源文。

[覺性 术语族扫描 / 禁固定译词审计]
如果用户挑战 覺性 译法，或审校触及 覺性 / 圓覺 / awakened-awareness 术语，必须扫描整个 artifact 与累积日志中的相关术语族：覺性、圓覺性、淨圓覺性、隨順覺性、圓覺、淨覺、覺知、本覺、始覺、佛性、自性、明、覺，以及相关上下文中的邻近复合词。建立或更新一致性矩阵，列出源词、本地源文语境、所选译法、否定的替代译法、理由，以及早期轮次是否已被取代。确认最终 artifact 没有机械固定译词、没有在需要时丢失 性/nature 力度、没有不必要的 “consciousness”、没有缺乏依据的 “buddha-nature”，也没有隐藏 Self / witness / pure-consciousness 漂移。

Prompt 6 / Prompt 9 / Strict QA 的 Batch89 扫描要求：
相关时必须扫描：覺性 / 圓覺性 / awakened-awareness 语域；非固定等值处理；Perfect Enlightenment 语境；buddha-nature 过度扩张；consciousness / pure-consciousness / witness 漂移；藏文/梵文语域见证误用；以及任何 覺性 挑战后的整族累积日志一致性。

强制 QA / changelog 确认行：
Batch89 覺性 语域关卡已检查：相关处已扫描 覺性 / 圓覺性 术语族；已选择上下文译法；没有固定等值替换；没有无依据的 “buddha-nature”；没有不必要的 “consciousness”；没有 Self/soul/pure-consciousness/witness 漂移；外部语域见证只作参考，未覆盖源文。

BATCH 88 — 菩提达摩术语族一致性 + 中文「體/受」语域 + 防反实体化过度削薄 加固 — 2026年7月6日

适用范围与继承：
本 Batch88 加固补丁适用于所有 Awakening to Reality / John Tan / Soh Wei Yu 翻译、审校、精修、对抗性审计、比较审计、累积日志、源文恢复、无修复晋级、prompt-manager、打包与交接工作流。它继承且不削弱 Batch76–87，包括 Self/self 语义大小写、术语库执行、source-role 分类、选择性合并、累积修复保留、视觉 QA、状态纪律与无修复晋级规则。

来自 Bodhidharma R37 比较审计的核心教训：
反实体化不能变成削薄源文。避免隐藏 Self、见证者、基底或分离的本质实体，并不等于删除 體、性、自性、本性、真如、法性、佛性、心、法身 等源文值。正确方法是在不实体化的英文、目标语表达或已记录的上下文译法中保留源文义值。

[术语政策挑战 / 一致性审计]
如果用户挑战某个术语选择，或外部语域核验改变/澄清了某个术语政策，必须先做术语族全局一致性审计，再继续无关审校。检查当前 artifact 与累积日志中受影响术语族的所有出现，包括旧修复、被否定的替代译法与相关复合词。每一项标记为：已对齐、已被后轮修复取代、需要修复、或有理由保留。若早期轮次使用了现在已否定的译法，必须确认当前 artifact 已不再保留它，除非明确记录理由。累积日志应追加或更新一致性矩阵：源词、当前译法、源文/上下文角色、政策状态、取代状态。

[中文 體-family 语域关卡]
对于中文佛教源词 體、體性、自體、本體、法體、道體、用體、聖體、別體、法界體、法界體性，不得用一个英文等值词全局机械替换。必须逐处判断。当 “the essence of X” 会暗示一个可分离的隐藏本质时，通常优先用 “X, in essence...”。当中文确实需要更强的 體 力度时，可用 “very essence”、“fundamental essence”、“intrinsic essence”、“Dharma-essence”、“sacred essence”等较强表达。體性 可在 體 与 性 都有语义作用时译为 “essence-nature”。“Dharma body” 只保留给 法身 / dharmakāya 或明确 身/body 语境，不得作为 法體 的自动译法。除非源文语境明确支持 body/embodiment，而不是 essence/constitution，否则避免把 體 译成裸 “body”。不得因反实体化而抹掉 體；应以不实体化的英文形式保留。若有意避免 essence 语言，必须记录为何那样会过度实体化，并确认源文 體 值没有丢失。

[反实体化，但不得削薄源文]
修复隐藏实体风险时，必须确认修复没有抹去 體、性、自性、本性、真如、法性、佛性、心、法身 等明确源文值。若修复后的译文删除或弱化了这些值，应以不实体化形式恢复，或记录为何上下文允许省略。不得把“删掉 essence 语言”当作默认反实体化策略；默认策略是上下文忠实、非实体化地保留。

[重复公式 / 平行段落扫描]
当修复高敏感公式、偈颂行、结尾偈、教义反复句或重复段落时，必须扫描全文是否有完全相同或近似重复。源文与语境相同者，应一致修复。若同一句/同一公式采用不同译法，必须记录上下文理由。不得假定一个局部修复已经修复了整个公式族。

[受 的上下文译法关卡]
受 必须按上下文译。业果、地狱经验、苦果、报应语境中，当 “receive” 会暗示有一个接收者实体或显得过于交易化时，优先考虑 “undergo”。教学、传授、有意接受或普通接收语境仍可用 “receive” 或 “accept”。若 受 明确是 vedanā，则用 “feeling/sensation”。若修复了 受，应扫描附近重复公式，如 無受、即受、誰作誰受，以及平行业果/地狱/苦果句。

[术语政策后的日志对齐审计]
任何轮次一旦宣布、修订或澄清一个术语政策，必须先审计累积日志，再转向无关问题。审计应列出：已对齐的改动、被取代的改动、未改但敏感的案例、剩余风险区域。若后轮修复取代了早期中间判断，早期日志必须明确标注“已被第 X 轮取代”，避免后续审校者把已修正的中间译法误当作当前政策。

[长迭代审计的经验总结 / 提示词加固报告]
对于长篇多轮审计，尤其用户询问应吸取什么经验时，最终或接近最终回应应包含“经验总结 / 提示词加固”报告。必须区分 artifact 实际改动与未来工作流改进。说明每类问题来自：规则缺失、规则过于含糊、已有规则未被操作化、源文歧义、过度修正、重复公式漏扫、累积日志漂移，或 public/sibling source 混淆。

Prompt 6 / Prompt 9 / Strict QA 的 Batch88 扫描要求：
必须扫描：术语政策挑战；受影响术语族一致性；中文 體-family 语域；反实体化但不削薄源文；重复公式 / 平行段落一致性；相关处的 受 上下文译法；术语政策后的累积日志对齐；日志中的取代标记。若发现其中一个缺陷，必须扫描整个 artifact 与累积日志中的同类缺陷。

强制 QA / changelog 确认行：
Batch88 术语族一致性关卡已检查：术语挑战响应、中文 體-family 语域、反实体化但不削薄源文、重复公式/平行段落扫描、受 的上下文译法、术语政策后日志对齐、取代标记，以及长审计经验报告（在相关处已应用）。

BATCH 86/87 COMBINED — SELF/self SEMANTIC-CAPITALIZATION + TERMBANK ENFORCEMENT + SELECTIVE-MERGE HARDENING — 2 July 2026

Scope and inheritance:
This combined universal hardening patch applies across all Awakening to Reality / John Tan / Soh Wei Yu translation, review, refinement, adversarial audit, formatting, Strict HTML QA, full-page integration, no-repair promotion, packaging, upload-controller, prompt-manager, handoff, source-restoration, and mechanical-cleanup workflows across all target languages and scripts. It does not weaken Batch76–85 safeguards, including href/src/code preservation, no-hidden-witness hardening, identity-syntax clarification, target-language-first technical terms, Total Exertion protection, Spectral/white-canvas/visual QA, or status discipline.

Important operational meta-rule:
If a termbank/doctrinal/source-role rule already exists in any prompt, the working assistant must not merely know it abstractly. It must operationalize it as a concrete artifact scan, source comparison, boundary check, or decision gate whenever the task touches that rule. Failure to apply an existing termbank or doctrinal rule is a material QA failure and must be logged as one of: rule absent; rule present but too vague; rule present but not operationalized; rule present and operationalized but missed due to source ambiguity; user preference changed after artifact; sibling-version conflict or merge-source confusion.

PATCH A / BATCH 86 — SELF/self SEMANTIC-CAPITALIZATION AND TRUE-SELF CRITIQUE GATE

Core lesson:
In AtR texts, English capitalization in “Self” is usually a semantic/doctrinal marker for a reified True Self / Big Self / Atman-like Self, especially when contrasted with lowercase “self” or when the article critiques “All as Self,” “No-Self based on Self,” “Self-Realization,” “True Self,” “Big Self,” “Great Self,” or “self/Self.” Do not mechanically translate the typography as “uppercase/lowercase self” in the target language unless the source sentence is explicitly about typography. The target-language rendering must carry the doctrinal contrast naturally while preserving the source label in parentheses where needed.

Mandatory Self/self decision gate:
Before finalizing any translation, merge, polishing, formatter pass, Prompt 6 review, Prompt 9 refinement, Strict HTML QA, or no-repair review, classify every article-central occurrence of: Self; self; self/Self; no-self/Self; True Self; Big Self; Great Self; Atman/Ātman; Self-Realization; Self-Inquiry / self-enquiry; All as Self; No-Self based on Self; Mine or Self. Classify each occurrence as ordinary lowercase self/ego/person-reference; capitalized Self as reified True Self/Big Self/Atman-like target of critique; protected AtR source label; exact title/quotation; ordinary English residue requiring translation; or ambiguous requiring gloss/human-review flag.

Anti-capitalization-calque rule:
The following are usually defects in non-English running prose unless the source sentence is literally about typography: local phrases meaning “uppercase Self,” “capital-letter Self,” “capitalized Self,” “lowercase self,” Japanese 大文字の自己 / 小文字の自己, Korean 대문자 자아 / 소문자 자아 / 대문자 Self / 소문자 self, or any equivalent where the target wording explains typography rather than doctrine. Do not translate source capitalization metadata as visible typographic commentary when the source is marking a True-Self/self-view contrast.

Preferred pattern:
Use a natural target-language doctrinal rendering plus the English source label in parentheses where useful. Lowercase self may be small self / ego / ordinary self or the natural local ordinary term. Capitalized Self may be True Self / Great Self / Big Self / Atman-like Self where this is the doctrinal target, plus “(Self)” when the English label matters. self/Self should become a natural local small-self vs True/Great-Self contrast, with “(self)/(Self)” if needed. Self-Realization should not become ordinary self-actualization when the source means realization of True Self.

Do not overcorrect:
Bare English “self/Self” can be acceptable as a protected source label, title fragment, exact quoted phrase, glossary label, or parenthetical after a target-language rendering. The defect is accidental residue, target-language grammar glued to English labels, or unnatural capitalization-calque that fails to convey the doctrinal contrast.

Language-specific Self/self examples:
Chinese: lowercase self → 小我 / 自我 depending context. Capitalized Self / True Self / Big Self → 真我 / 大我 where the source critiques or names that view. self/Self → 小我／真我 or 小我／大我, preserving “(self)/(Self)” where useful. Avoid literal 大写的Self / 小写的self unless the sentence is actually about typography.
Japanese: lowercase self → 小我（self） or 自己/自我 where ordinary self is meant. Capitalized Self as True/Big Self target → 真我・大我（Self）; 真我（True Self） for explicit True Self; 大我 or 大いなる自己 for explicit Great/Big Self when context supports it. self/Self → 小我（self）／真我・大我（Self）. Avoid 大文字の自己 / 小文字の自己 for doctrinal Self/self contrast.
Korean: lowercase self → 소아(self) or 자아(self) depending context; prefer 소아 where explicitly contrasted with capitalized Self. Capitalized Self / True Self / Big Self → 진아·대아(Self), 진아(True Self), 대아(Great/Big Self). 참나 may be legitimate in Ramana/True Self contexts only when context and house style allow it. self/Self → 소아(self)와 진아·대아(Self). Avoid 대문자 자아 / 소문자 자아 / 대문자 Self / 소문자 self for doctrinal contrast.
For all other languages: choose a language-specific doctrinal contrast. If stable local Buddhist/Vedantic terminology exists for True Self / Big Self / Atman-like Self and the source is critiquing that view, use the natural term plus the English source label as needed. If no safe local term is known, preserve “Self” as a protected label with a concise target-language explanation instead of inventing a capitalization-calque. Do not force Chinese/Japanese/Korean terms onto unrelated languages.

Prompt 6 / Prompt 9 / Strict QA scan requirements for Self/self:
Mandatory scan for capitalized Self / lowercase self / self/Self / no-self/Self; local phrases meaning uppercase/lowercase Self/self; target-language grammar attached to English self/Self labels; Self-Realization mistranslated as ordinary self-actualization; True Self / Big Self / Great Self neutralized when the source is explicitly critiquing that view; bare self/Self source labels that may be acceptable but must be classified. If one capitalization-calque or Self/self defect is found, scan the entire artifact and sibling files in the batch for the same defect family.

PATCH B / BATCH 87 — TERMBANK ENFORCEMENT / SELECTIVE-MERGE / REFUGE-REALIZATION-SPONTANEOUS-PERFECTION HARDENING

Mandatory source-role classification gate:
At the start of any task involving multiple artifacts, older fuller pages, newer partial translations, live/public web pages, appended materials, or “latest artifact” workflows, classify each source internally before editing: (1) English original / doctrinal source authority; (2) current working artifact / latest cumulative merged artifact; (3) newer partial translation/design artifact; (4) older fuller page artifact; (5) live/public AtR page; (6) appended/subsequent articles after the main target article; (7) user-pasted correction or user-stated term preference. Do not let an older public translation override the English source or the latest attached translation wording when the user wants updated translation wording. Do not lose later repairs merely because a sibling file lacks them.

Do not wholesale replace unless explicitly requested:
When a user provides an updated sibling file and says “apply latest version changes” or “take note what latest version changed,” do not automatically copy that sibling file wholesale into the current artifact. First determine whether the user wants wholesale replacement of a defined region, selective application of relevant fixes, design-only transfer, text-only transfer, termbank-only transfer, or comparison report only. If the user later clarifies not to copy wholesale, roll back to the last safer cumulative base and selectively reapply relevant fixes.

Cumulative repair preservation rule:
Never assume the highest-numbered or newly attached file includes all prior repairs. Before using it as a source, compare it against the current cumulative artifact and check for regressions in user-reported fixes, doctrinal terminology repairs, link and URL repairs, structural/full-page preservation, HTML/CSS repairs, iframe/title/target/rel safety repairs, source-text fidelity repairs, and prompt-specific termbank decisions. If a sibling file lacks a later repair, do not reintroduce the older defect.

Chinese realization terminology gate:
For Chinese AtR / John Tan / Soh articles, use 证悟 as the default for major direct realization-language: realization; realized; realize anatta; I AM realization; realization of emptiness; Self-Realization; realization vs experience; if you think you realized Stage 5; realization of the view; direct realization; awakening/realization stages. Use 了悟 only when the meaning is closer to “understood clearly / came to understand / saw the point,” or in inherited titles, legacy quotes, poems, fixed Chinese source material, or contexts where 了悟 is intentionally chosen for tone. If 了悟 is found in a major direct-realization context inside the active article core, treat it as a terminology risk and decide whether it should be 证悟. Do not apply 了悟 → 证悟 blindly across appended legacy articles or quoted Chinese source material unless the user asks for global modernization.

Cross-language realization / experience / understanding gate:
For all languages, distinguish direct realization / awakening / experiential verification / irreversible insight; temporary experience / glimpse / recognition; conceptual understanding; comprehension of a point; attainment / fruition; meditative state or absorption. Do not flatten realization into mere understanding when the source describes a major insight-realization. Do not translate experience as realization, and do not translate realization as mere experience. If a defect is found, scan the whole artifact and sibling files for the same defect family.

Spontaneous Perfection gate:
For Chinese AtR Seven Stages / On Anatta / Stage 7 contexts, prefer 自然圆成 as the technical house rendering in the active translated article core unless the user or a fixed title requires otherwise. Spontaneous Perfection → 自然圆成; Presence is Spontaneously Perfected → 临在是自然圆成的; On Spontaneous Perfection → 论自然圆成. “already perfect and uncontrived nature” may be rendered naturally, often 本已圆满、无造作的本性, but do not confuse this with the technical title/section term Spontaneous Perfection. Do not globally rewrite every occurrence of 自然圆满 in appended legacy materials, already-published older Chinese article titles, quoted Chinese materials, or unrelated sections unless the user asks for global harmonization. For all languages, identify Spontaneous Perfection as a technical Stage 7 / Dzogchen-adjacent / AtR technical term, not ordinary moral or generic perfection; use a consistent article-specific rendering and preserve the English term in parentheses if needed.

Stage 5 aggregate-emptying logic gate:
The Seven Stages sentence “The practitioner will also be naturally led into spontaneous perfection without going further in emptying the aggregates” must not be translated as if “there is no need to empty the aggregates,” because the immediately following comment says there is a need to further empty the aggregates and eighteen dhātus. Preserve the provisional/diagnostic meaning: the practitioner may be led into spontaneous perfection while not yet having further emptied the aggregates; this does not cancel the need stated in the next paragraph. Chinese preferred: 修行者也会自然而然地被引向自然圆成，而尚未进一步空掉五蕴。 Acceptable: 修行者也会自然而然地被引向自然圆成，即便尚未进一步空掉五蕴。 Avoid: 修行者也会被自然地引导进入自然圆满，而不需要在空掉五蕴方面走得更远。

Refuge / taking refuge gate:
If the source says refuge, take refuge, sought refuge, go for refuge, or links to Taking Refuge / Three Jewels / Buddha-Dharma-Sangha material, classify whether it is Buddhist refuge in the Three Jewels, general religious refuge, metaphorical shelter, secular safety/shelter, or exact title/link label. Use the target language’s Buddhist refuge term when the source is Buddhist refuge. Do not flatten it into secular safe place / shelter / harbor / 避风港 unless the context is clearly secular. Chinese: 皈依 / 归依; 皈依三宝 / 归依三宝; 皈依佛教 / 在佛教中寻求皈依; avoid 在佛教中寻求避风港 when doctrinal. Japanese: 帰依; 三宝に帰依する; avoid 避難所 / 安息所 when doctrinal. Korean: 귀의 / 귀의하다; 삼보에 귀의하다; avoid 피난처 / 안식처 when doctrinal. For other languages, use established local Buddhist refuge terminology if available; if uncertain, preserve “take refuge in Buddhism / the Three Jewels” rather than secularizing it.

Termbank enforcement and prompt-compliance checkpoint:
Before finalizing, identify whether the article touches any known prompt/termbank rule family, including realization vs experience vs understanding; Self/self/True Self/Big Self; Witness/sākṣī/no-hidden-witness; agent/doer/doership; nondual/non-dual; no-self/anatta/anatman; emptiness/śūnyatā; dependent origination; spontaneous perfection; Buddha Nature; refuge/taking refuge; consciousness/awareness/mind; rigpa/vidyā; Presence; I AM; ordinary/ordinariness; Maha; Total Exertion; source-title/link-label preservation; Chinese 體/體性/自體/本體/法體/道體/用體/聖體/別體/法界體/法界體性; 受; repeated formulas / parallel passages; and term-policy conformance challenges. For every touched family, run a targeted scan or sampling check. If the prompt already contains a rule and the artifact violates it, state whether this was a rule absence, rule vagueness, non-operationalization, source ambiguity, post-artifact preference change, or sibling-version/merge confusion.

Selective full-page integration gate:
When integrating an updated translation/design into an older fuller page, preserve and distinguish: the newer translated main article core; older top opening/resource block; correct language URLs and related links; audio embeds and media cards; original English link; preface/intro blocks; table of contents; post-main-article Soh commentary; appended/subsequent articles after the main article; sidebar/library content inside the exported post body; HTML/CSS design blocks required by the latest design prompt; and any previous repairs made after the newer translation/design file was produced. Do not delete full-page material when transplanting a new translation core. Conversely, do not assume every new translation core should overwrite post-core appended materials.

Boundary gate:
Define integration boundaries explicitly. The “main article core” begins at the main article title/stage section or first stage heading, depending source structure, and ends after the main article’s final stage and immediate commentary/P.S. if included. Appended/subsequent materials begin at the next distinct article heading or article boundary. If uncertain, use headings, article titles, anchor structure, and source comparison. Do not move a correction across boundaries without deciding whether it belongs to the main core or appended tail.

Main-core versus appended-tail scope:
If the user says “update the Seven Stages Chinese” or “apply latest Seven Stages fixes,” default to updating the Seven Stages main article core only. Do not rewrite appended/subsequent articles unless the user explicitly requests global update across appended materials. If a high-risk term defect appears in appended material too, report it separately as fixed in main core, left untouched as appended legacy material outside scope, or recommended for a separate global harmonization pass.

No-regression artifact version rule:
When producing Rxx versions, the new version must preserve all repairs from prior versions unless explicitly reverted. Include a preserved-repairs checklist in the audit log: latest intended translation/design retained; no lost 皈依 repair; no regression to 避风港; no regression to 自然圆满 in main core when 自然圆成 is intended; no regression to 了悟 in major direct-realization contexts where 证悟 is intended; no regression to risky Stage 5 “不需要” wording; no deletion of requested opening/resource block; no deletion of appended articles; no replacement of newer partial text over cumulative repairs; no link/src/href loss; no iframe/title/rel safety loss; no raw HTML/CSS leak.

Prompt 6 / Prompt 9 mandatory review baskets:
Every high-fidelity review and whole-document refinement pass must include: termbank enforcement basket; realization/experience/understanding basket; Self/self semantic-capitalization basket; refuge basket; Spontaneous Perfection basket; aggregate-emptying logic basket; source-role basket; selective-merge basket; cumulative repair basket; boundary basket; public-page caution basket; no-repair promotion basket; Batch88 terminology-policy challenge / conformance basket; Chinese 體-family register basket; anti-substantialism-without-source-thinning basket; repeated formula / parallel passage basket; 受 contextual-rendering basket; and post-policy log-alignment basket when a term policy has been declared, revised, or challenged.

Strict HTML QA / Formatter / Mechanical utility application:
Strict HTML QA and formatting prompts must not only check HTML mechanics. When text is being rewritten, translated, merged, normalized, or mechanically cleaned, include a limited material-risk scan for high-risk visible terminology. Formatting tools must preserve 证悟 where intentionally used, 皈依/Buddhist refuge terminology, 自然圆成 in technical Spontaneous Perfection contexts, target-language Self/self glosses and English source labels, source-title/link-label distinctions, doctrinal parentheticals, and main-core/appended-tail boundaries. They must not normalize chosen target-language terms back into older public-page wording, perform global replacement across appended articles without scope authorization, treat a newer sibling file as whole replacement merely because it has a higher version number, or remove intentional source labels such as (Self), (self), True Self, I AM, Presence, rigpa, or anatta.

Status discipline:
If a terminology, source-fidelity, merge-boundary, source-role, cumulative-repair, or doctrinal-logic repair is made, status must be: REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required. Do not promote in the same pass. Only a later exact-artifact no-material-edit review may use: reviewed final / publishable — strict-certified final not claimed. When a pass only creates an audited copy without material change, the log must explicitly say no material changes were made and that prior repairs remain preserved.

Mandatory QA/changelog confirmation line:
Batch86/87 Self/self + termbank/selective-merge gates checked: Self/self capitalization-calque scan, realization-vs-experience scan, Buddhist refuge scan, Spontaneous Perfection scan, Stage 5 aggregate-emptying logic scan, source-role classification, selective-merge boundary check, cumulative repair preservation, and no-regression review completed.
Batch88 term-family conformance gates checked: terminology challenge response, Chinese 體-family register, anti-substantialism without source-thinning, repeated formula/parallel passage scan, 受 contextual rendering, post-policy log-alignment audit, supersession marking, and long-audit lessons report applied where relevant.

BATCH 85 — WHITE CANVAS SUPREMACY / AUTHORIAL-PROSE DEFAULT / QUOTE-CITATION HIERARCHY / LONG-TITLE CONTROL — 28 June 2026

This Batch85 patch applies to live Blogger/HTML teaching-article formatting, Strict HTML QA, formatter readback, visual audit, final audit, packaging, prompt-manager publication, standalone TXT prompt packages, public standalone HTML pages, copyable embedded prompt bodies, Blogger replacement bodies, manifests, changelogs, handoff files, and upload/publishing/controller workflows. It applies across all target languages and all target scripts. It is a visual-formatting and HTML-display hardening patch. It does not permit omission, paraphrase, doctrinal rewriting, source-link mutation, source-title mutation, translation change, or quotation/content alteration.

Routing note for prompts that do not themselves create or audit styled HTML: if the workflow can lead to a Blogger/HTML artifact, route any visual-formatting issue through the Unified Blogger Formatter plus Strict HTML QA before publication.

1. WHITE CANVAS SUPREMACY FOR LIVE BLOGGER TEACHING ARTICLES
For live Blogger teaching articles, the page/post canvas behind all ordinary article content must remain clean white all the way down unless the user explicitly asks for a dark/dawn poster layout. The default root wrapper and common Blogger parent containers must not show yellow, cream, lavender, dawn, radial, or linear gradient washes behind the whole post or behind ordinary prose.

Required behavior:
- The article root must be an invisible layout container on a white canvas, not a visible outer showcase shell.
- Apply final-winning rules such as `background: #fff` and `background-image: none` to the article root and common Blogger post containers where safe.
- Do not put gradients behind all boxes, all prose cards, or the whole article body.
- Gradients are allowed only for clearly intentional components: hero banners, dark formula boxes, explicitly requested Plum Dawn feature panels, or ornamental dividers. They must not become the article-wide background.
- Ordinary prose, notes, source cards, quote cards, resource cards, language panels, and table/list sections should use white or very subtle cool-lavender/blue/white fills, not warm/yellow gradients.

Required audit lines:
- White canvas behind all ordinary content: PASS/FAIL.
- Article/root background-image gradient residue: [count/list].
- Ordinary card gradient residue: [count/list].
- Gradients limited to intentional hero/dark/plum components: PASS/FAIL/NOT APPLICABLE.
- Blogger parent white-canvas override present and final-winning: PASS/FAIL.

2. AUTHORIAL-PROSE DEFAULT: DO NOT BOX MAIN NARRATION JUST BECAUSE THE AUTHOR IS JOHN TAN / THUSNESS / PASSERBY / SOH
For teaching articles, the main authorial narration should normally remain plain flowing prose on the white canvas. Do not automatically wrap every John Tan / Thusness / PasserBy / Soh paragraph in visible cards merely because the speaker is important.

Required behavior:
- Main authorial exposition defaults to normal paragraphs, with clean spacing and Spectral typography.
- Use visible boxes for actual source quotations, blockquotes, stanza/poem passages, reader questions, dialogue turns, Soh translator/editor notes, resource/download/list/meta sections, and real post/comment boundaries.
- Do not make the main body look like a sequence of undifferentiated boxes when the source article is continuous authorial teaching.
- “Logical boxing” means grouping real units, not decorating all paragraphs.
- If the source contains continuous multi-paragraph exposition by one author, keep it as flowing prose unless there is a real speaker/source/topic boundary.

Required audit lines:
- Main authorial prose left as readable flowing prose: PASS/FAIL.
- Important author name did not trigger automatic boxing: PASS/FAIL.
- Paragraph-as-card overdecorating scan: PASS/FAIL.
- Continuous teaching passage sample checked: [number].

3. QUOTE / CITATION / NOTE HIERARCHY MUST BE VISUALLY DISTINCT
The visual system must make it immediately clear what is main narration, what is a quotation/citation, what is a stanza/poem, and what is an editor/Soh note.

Required behavior:
- Actual blockquotes and cited excerpts use a distinct quote-card/source-quote style.
- Stanzas/poems use a stanza-card style and may preserve meaningful line breaks.
- Soh/editor comments use a note/commentary style distinct from source quotations.
- Top meta/resource/audio/language cards use meta/resource styles distinct from main article quotations.
- Quote labels such as “~ Thusness” must remain visually tied to the quote and not be confused with normal paragraph endings.
- If all major text blocks look equally boxed, fail visual hierarchy.

Required audit lines:
- Main prose vs quote/citation hierarchy: PASS/FAIL.
- Source quotes visually distinct from authorial narration: PASS/FAIL.
- Stanza/poem styling distinct and line breaks preserved: PASS/FAIL/NOT APPLICABLE.
- Soh/editor notes visually distinct from source quotes: PASS/FAIL/NOT APPLICABLE.
- Quote labels remain attached and readable: PASS/FAIL.

4. LONG TITLE CONTROL / NO POSTER-SIZED H1 FOR LONG ARTICLE TITLES
Big cinematic titles are allowed, but long article titles must not become oversized poster text. The formatter must use length-aware and script-aware title sizing.

Required behavior:
- If an English/Latin h1 is roughly 55+ characters, or if any title wraps to more than two lines at common desktop widths, apply a long-title class or equivalent CSS.
- Suggested long-title cap: `clamp()` around 1.9rem–3.25rem, with max not exceeding about 3.35rem unless the title is short.
- Avoid max title sizes above 3.45rem for long titles.
- Use tighter but readable line-height around 1.08–1.16 for long titles.
- For CJK, Tibetan, Thai, Arabic, Devanagari, Tamil, Bengali, and other scripts, use rendered-line behavior rather than raw English character count when possible. If browser/rendered check is unavailable, use conservative caps for long or wrapped titles.
- If the source/original title was modestly sized and the new design greatly enlarges it, require explicit justification.

Required audit lines:
- Long-title detection applied: PASS/FAIL/NOT APPLICABLE.
- H1 max-size appropriate for title length/script: PASS/FAIL.
- Title wraps without crowding/squeezing: PASS/FAIL.
- No oversized poster-title regression: PASS/FAIL.

5. VISUAL QA MUST INCORPORATE USER-REPORTED SCREENSHOT/READABILITY DEFECTS
If the user reports that something looks wrong in preview — too large title, non-white background, overboxing, indistinct quote hierarchy, dark-on-dark text, narrowed canvas, hard linewraps, or similar — that report becomes a mandatory risk-basket item in the next audit pass.

Required behavior:
- Do not pass by text parity alone.
- Reopen the exact latest artifact and inspect CSS/DOM/class roles against the reported visual defect.
- If screenshot/browser computed style is unavailable, state that limitation and perform static CSS/DOM checks; do not pretend rendered review happened.
- A defect repaired through CSS/DOM/layout changes is material and cannot be promoted in the same pass.

Required audit lines:
- User-reported visual defects incorporated: PASS/FAIL/NOT APPLICABLE.
- CSS/DOM evidence for each visual defect: [brief list].
- Text/link parity was not substituted for visual QA: PASS/FAIL.
- Browser/screenshot review performed: PASS/FAIL/NOT AVAILABLE.

6. IFRAME ACCESSIBILITY AND BLOGGER MACRO EXCEPTION CLARIFICATION
Keep the existing Blogger native video macro exception: do not add title attributes or modify attributes on `iframe.BLOG_video_class`. For ordinary third-party iframes that are not Blogger native macros, such as SoundCloud or standard external embeds, a descriptive `title` attribute may be added if missing, while preserving `src` and other embed-critical values exactly.

Required audit lines:
- Non-Blogger iframe title/accessibility check: PASS/FAIL/NOT APPLICABLE.
- Blogger native video macro exception respected: PASS/FAIL/NOT APPLICABLE.
- Iframe src/embed-critical attributes preserved: PASS/FAIL.

7. STATUS DISCIPLINE
Any repair to white canvas/backgrounds, gradient removal, authorial-prose boxing, quote hierarchy, long-title CSS, class-role separation, iframe accessibility, or visual-layout behavior is material. Use:

REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required.

Only a later exact-artifact no-material-edit pass may use:

reviewed final / publishable — strict-certified final not claimed.

8. PACKAGING / PUBLIC PROMPT SYNC
After applying this patch, update the live prompt source files, standalone TXT prompt packages, public standalone HTML pages, copyable embedded prompt bodies, Blogger replacement bodies where present, current-use labels, manifests, changelogs, handoff files, and any prompt-manager package index. Confirm that embedded public HTML prompt bodies and standalone TXT prompt bodies are synchronized.

Batch85 session lessons to preserve in changelog and future review:
- R2 exposed missing ordinary non-Blogger iframe title handling.
- R3 exposed long-title oversizing despite generic “big but controlled” language.
- R4 exposed ambiguity between “logical boxing / balanced card density” and the need to preserve plain authorial narration plus distinct quote/citation hierarchy.
- R4 confirmed that “white canvas” must mean no article-wide or ordinary-card gradient residue; gradients should be limited to intentional hero/dark/plum components.
- Future audits must not rely on normalized text/href parity alone when the user is reporting visual readability defects.

Mandatory Batch85 QA line:
- Batch85 visual logic / white canvas / quote hierarchy / long-title gate checked: white canvas remains dominant, main authorial prose is not over-boxed, quote/citation/note hierarchy is distinct, long titles are controlled, user-reported visual defects were incorporated, and Blogger iframe macro exceptions were respected.

BATCH 84 — LOGICAL BOXING / LINEWRAP CLEANUP / SPECTRAL ENFORCEMENT / BLOGGER VISUAL READBACK HARDENING — 25 June 2026

This Batch84 patch applies to Blogger formatting, Strict HTML QA, formatter readback, final audit, packaging, prompt-manager publication, standalone TXT prompt packages, public standalone HTML pages, copyable embedded prompt bodies, Blogger replacement bodies, manifests, and changelogs. It is a formatting/display hardening patch. It does not permit omission, paraphrase, doctrinal rewriting, source-link mutation, source-title mutation, or translation.

1. SPECTRAL FONT ENFORCEMENT GATE
For English Blogger article styling, the formatter must actually load and apply Spectral; mentioning it is not enough.

Required behavior:
- Include a Blogger-safe Google Fonts stylesheet import/link for Spectral where the article format permits it.
- Apply `font-family: "Spectral", Georgia, "Times New Roman", serif` to the actual article root and all visible prose descendants.
- Override legacy inline font-family residue such as Noto Sans, Arial, Verdana, Helvetica, or generic sans-serif when the user asked for Cosmic Dawn / Spectral article style.
- Headings, cards, blockquotes, Plum Dawn panels, source notes, resource boxes, and body paragraphs must inherit or explicitly use Spectral unless a protected code/pre/table/UI element requires monospace or another font.
- Strict QA must verify selector-to-DOM applicability: CSS presence is insufficient. The actual wrapper/content selectors must match the generated DOM.

Required audit lines:
- Spectral font loaded: PASS/FAIL.
- Spectral applied to actual article root: PASS/FAIL.
- Legacy body font-family residue overridden: PASS/FAIL.
- Selector-to-DOM font applicability: PASS/FAIL.

2. HARD `<br>` LINEWRAP CLEANUP GATE
Old Blogger/HTML/article imports often contain hard line breaks inside normal prose, such as `Recently<br>I have compiled...` or `Thusness commented in his 7 stages<br>article that...`. These are export line wraps, not meaningful paragraph breaks.

Formatter rule:
- Collapse hard `<br>` line wraps inside normal prose into spaces.
- Preserve true paragraph breaks as separate paragraphs.
- Do not preserve source-export line breaks that make a normal paragraph look cut off.
- Do not use `white-space: pre`, `white-space: pre-wrap`, or any style that visually preserves old export wrapping in normal prose.
- Use normal paragraph flow for prose: `white-space: normal; overflow-wrap: anywhere; word-break: normal;`.
- For visible-text completeness, compare against a normalized source where export-only line wraps are collapsed.
- Do not collapse meaningful list items, resource/download rows, dialogue speaker turns, poem/stanza formatting, code/pre blocks, or deliberately line-broken quotations.

Strict QA must:
- Scan for `<br>` inside normal `<p>`, `.content-card`, `.source-quote`, `.atr-note`, `.plum-dawn-panel`, and normal prose cards.
- Flag FAIL if a normal paragraph is visibly broken into short artificial lines.
- Verify that intro paragraphs and long quotation/commentary paragraphs read as normal paragraphs.
- Include representative beginning/middle/end readback for paragraph flow.

Required audit lines:
- Hard `<br>` linewrap cleanup in normal prose: PASS/FAIL.
- Normal prose `<br>` residue count: [number].
- Paragraph reflow spot-checks: PASS/FAIL.
- Meaningful line breaks preserved where required: PASS/FAIL.

3. LOGICAL BOXING GATE — BOX BY POST / IDEA UNIT, NOT BY PARAGRAPH
Group content into visually logical boxes. Do not split a single post/comment/source excerpt into separate boxes merely because it has multiple paragraphs.

Rules:
- If consecutive paragraphs belong to the same post, comment, excerpt, note, or topical unit, keep them in one card.
- Paragraphing inside a card is allowed and often required.
- Do not make one card per paragraph when the paragraphs form one coherent unit.
- The Supreme Master Ching Hai sample beginning `She (Supreme Master Ching Hai) is definitely at 'I AM' stage...` plus `For one who realises non-duality...` must remain one logical card when present.
- Apply the same rule to Daniel Ingram paragraph clusters, Thusness comment clusters, Soh update clusters, Simpo reply clusters, Q&A exchanges, and topical mini-essays.
- Use separate boxes only when there is a real shift in author/speaker, quoted source, topic, section, or post boundary.

Required audit lines:
- Logical single-post boxing: PASS/FAIL.
- Paragraph-as-card over-splitting scan: PASS/FAIL.
- Multi-paragraph card samples checked: [number].
- Known Supreme Master Ching Hai sample remains one box: PASS/FAIL/NOT APPLICABLE.

4. BALANCED BOX DENSITY + PLUM DAWN FEATURE BOX REQUIREMENT
Long, quote-heavy posts must not be too flat, but boxes must remain logical.

Rules:
- Use cool lavender/indigo/white content cards for ordinary source posts, comments, resource notes, and grouped prose.
- Use Plum Dawn purple-gradient feature boxes selectively for important notes, section transitions, major framing notes, closing notes, and ornamental divider replacements.
- Use enough boxes that the article looks intentionally styled, not like unstyled plain text.
- Avoid a giant visible outer showcase/card shell inside Blogger. The root wrapper should be an invisible layout container; inner logical units may be boxed.
- Avoid nested Plum Dawn panels.
- Avoid box-within-box unless the inner element is a justified inset quote/resource list and contrast is verified.
- Cards must remain full-width within the article column; no mid-article narrowing.

Required audit lines:
- Balanced box density: PASS/FAIL.
- Content-card count: [number].
- Plum Dawn feature panel count: [number].
- Nested-card / box-within-box problems: PASS/FAIL.
- Mid-article width consistency: PASS/FAIL.
- Root wrapper invisible / not a visible showcase shell: PASS/FAIL.

5. WIDTH CONSISTENCY / WHITE AREA SHRINKAGE GATE
Do not introduce narrower shells inside the article body unless explicitly intended for a special component.

Rules:
- Do not set random `max-width: 760px`, `820px`, `900px`, etc. on mid-article prose cards.
- Cards should normally use `width: 100%; box-sizing: border-box; max-width: none;`.
- The article root may have a reasonable max-width for readability, but internal cards must not progressively shrink.
- Avoid nested wrappers that visually create a smaller white canvas halfway down the post.

Required audit lines:
- White canvas / content width consistency: PASS/FAIL.
- Restrictive mid-article width rules found: none / [list].
- Internal cards full-width within article column: PASS/FAIL.

6. SEPARATOR TEXT CLEANUP GATE
Standalone separator-only lines such as `-----------------`, `...`, `....`, `—-`, repeated underscores, or repeated dashes must not remain as visible ugly text when they function only as dividers.

Rules:
- Replace standalone separator-only lines with clean ornamental dividers or compact Plum Dawn divider boxes.
- Do not remove real sentence ellipses such as `Yep...` or `This part described...` when they are part of prose.
- Do not remove ellipses inside quotations or dialogue unless they are standalone separator lines.
- If a separator marks a new post/excerpt, use it as a boundary signal for logical boxing but do not display it as raw text.

Required audit lines:
- Standalone separator text residue: PASS/FAIL.
- Separator-only lines converted to styled dividers: PASS/FAIL.
- Lexical sentence ellipses preserved: PASS/FAIL.

7. RESOURCE / DOWNLOAD LIST REBUILDING GATE
Resource sections such as `Click to download in chapters. Preface / How to use the articles / Who are we? ...` must not remain as flat broken paragraphs.

Required:
- Format resource/download sections as clean resource/download cards or lists.
- Keep all href values exactly.
- Keep resource labels and descriptions.
- Normalize safe spacing such as `Who are we?(113kb)` to `Who are we? (113kb)` without changing meaning.

Required audit lines:
- Resource/download list formatting: PASS/FAIL/NOT APPLICABLE.
- Resource href parity: PASS/FAIL.
- Resource labels/descriptions preserved: PASS/FAIL.

8. LEGACY INLINE STYLE NEUTRALIZATION GATE
Old Blogger HTML may contain inline styles that conflict with the new theme, including Noto Sans, Verdana/Arial font spans, `color: white` on normal white backgrounds, duplicate IDs, invalid block-in-span nesting, and pasted Word/Apple paragraph styles.

Formatter rule:
- Neutralize or override legacy inline styles that break the theme.
- Preserve meaning, links, images, and visible text.
- Remove or repair duplicate IDs.
- Repair invalid block-in-inline structures.
- Force readable contrast in all cards/panels.

Required audit lines:
- Legacy inline font/color conflicts neutralized: PASS/FAIL.
- Duplicate IDs: [count].
- Invalid block-inside-inline nesting: PASS/FAIL.
- White-on-white / dark-on-dark text risk: PASS/FAIL.

9. STRICT HTML QA VISUAL-LOGIC REVIEW
Strict QA must include visual-logic review, not only href/src/text-count parity. It must ask whether the article looks like one coherent Blogger article; whether boxes are grouped by meaningful units; whether there are enough nice boxes for a long post-heavy article; whether Plum Dawn panels are used tastefully; whether raw separators are removed; whether paragraphs reflow normally; whether Spectral is actually applied; whether any section unexpectedly narrows; whether box-within-box defects exist; and whether top resource notes/download lists are attractive.

Required audit lines:
- Visual-logic styling review: PASS/FAIL.
- Real Blogger preview risk items: none / [list].
- Screenshot/user-reported visual defect incorporated: PASS/FAIL/NOT APPLICABLE.
- Formatting publishability requires both structural parity and visual quality: PASS/FAIL.

10. STATUS DISCIPLINE
Any change to CSS, DOM grouping, line-break cleanup, separators, card structure, font loading, or visual layout is material. Same-pass publishable promotion is forbidden after such repairs.

Use exactly:
REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required.

Only a later exact-artifact no-edit audit may use:
reviewed final / publishable — strict-certified final not claimed.

11. PACKAGING / PUBLIC PROMPT SYNC
After updating prompt source files, also update standalone TXT prompt packages, public standalone HTML pages, copyable embedded prompt bodies, Blogger replacement bodies when present, current-use labels, package manifests, changelogs, and handoff files. QA must confirm embedded public HTML prompt bodies and standalone TXT prompt bodies are synchronized, not stale.

Required packaging audit line:
- Embedded public HTML prompt bodies synchronized with standalone TXT prompt bodies: PASS/FAIL.

BATCH 83 — 目标语优先技术术语 / Total Exertion 保护 / QA 结构加固 — 2026年6月23日

本 Batch83 补丁适用于完整 AtR 翻译、审校、Prompt 1 严格 HTML 保留、Prompt 6 对抗性审校、Prompt 9 最终/可发布 QA、各语言专门关卡、formatter prompts、upload/publishing/controller prompts、prompt-manager 流程、术语库 prompts、链接处理、对话保留与最终 artifact 晋级。它适用于所有目标语言，不只适用于阿拉伯语。

1. 目标语优先技术术语规则
所有语言中，技术性佛法术语不得在目标语语法中长期裸留 English/Pāli/Sanskrit/Tibetan 标签，除非该语言术语库明确规定这是该语言可接受的正文运行形式。默认使用目标语运行术语；必要时，尤其在首次出现、标题、或源术语具有教义重要性处，才在括号中保留源标签。不得把 English/Pāli/Sanskrit 标签机械粘入目标语前后缀或普通语法中。

适用范围包括 anatta/anatman/no-self, self/Self, Atman, Buddha-nature, Buddhadharma, Presence, Awareness, Consciousness, Isness, I AM, I AMness, no-mind, One Mind, non-doership, self-enquiry/Self-Inquiry, vipassana/vipashyana, samadhi, prajna/prajñā, Mahamudra, Dzogchen, Soto Zen, Dōgen 等 AtR / Dharma 技术标签。各语言术语库可以规定默认形式是翻译、音译或混合，但必须自然地融入目标语语法。

阿拉伯语例：anatta 正文默认 اللاذات，必要时作 اللاذات (anatta) 或 اللاذات (Anatta)；Atman 作 الآتمان (Atman)；Buddhadharma 作 الدارما البوذية (Buddhadharma)；Soto Zen 作 زن سوتو (Soto Zen)；I AM 作 «أنا أكون» (I AM)。避免把裸 anatta、Atman、Buddhadharma、Soto Zen、I AM 机械塞进普通阿拉伯语语法。

2. Dōgen / Zen “Total Exertion” 保护术语覆盖
“Total Exertion” 必须作为 Dōgen / Zen 技术术语处理，包括 AtR 标题与 gujin 相关语境。不得用普通词典义译成耗尽、枯竭、消耗、排空、过度用力，或仅仅个人努力/投入。若语言专门术语库已有批准译法，则使用该译法；若无，则保留 English technical term，并可按该语言规范音译。

阿拉伯语 artifact/default 批准形式：توتال إكسيرشن (Total Exertion)。不得使用 الاستنفاذ الكلي，因为它表示耗尽/枯竭。除非用户明确要求，不要自动添加首次解释说明；若项目专门 prompt 要求说明，必须简短，不得扩写或加编者解释。

3. 成语/语境对抗性关卡
Prompt 6 与 Prompt 9 必须检查看似简单的英文短语是否按语境翻译，而不是逐词照搬。在心理/灵性证悟语境中，saw the same thing 往往不是“看见同一个物理对象”，而是“理解/体验/认出同一个体验或洞见”。

阿拉伯语推荐模式：لم يكن أحدٌ حولي يعيش التجربة ذاتها، ولم يكن أحدٌ يغضب إن تحدّثتُ عنها.

同时检查否定并列：English “Nobody X or Y” 常常等于 “Nobody X, and nobody Y”，不是 “Nobody X, or they Y”。不得生成尴尬的“或他们生气”式结构。

4. 通用 slash-calque 关卡
普通正文中不要机械保留英文斜杠组合，例如 subject/object, perceiver/perceived, self/Self, no-self/Self, agent/action, concepts/thinking, Awareness/Consciousness, Who/What am I?, Soto Zen/Dōgen, Eight Negations / Middle Way negation。应使用目标语自然并列，如“和”“或”“X 与 Y 之间”，或括号保留源标签。只有当斜杠属于精确标题、精确公式、文件名、代码、URL、或受保护技术标签时才保留。

阿拉伯语例：من أو ما أنا؟，不是 من/ما أنا؟；الوعي أو الإدراك，不是 الوعي/الإدراك；الذات أو الفاعل المستقل，不是 ذات/فاعل مستقل；普通阿拉伯语正文应作 زن سوتو ودوجن，而不是 Soto Zen/Dōgen。

5. 链接所在句 QA
每个 anchor tag 都必须保持 href 精确不变，除非用户明确要求改链接。必须保持正确 anchor 边界：链接文字不得吞掉邻近句子文字、标点或相邻链接。若源链接文字是英文文章标题而目标 artifact 翻译链接文字，整句仍必须自然。若精确标题必须保留英文，不得生硬粘进目标语语法，需用自然框架嵌入。

对抗性检查：读完整个含链接句子，不只读链接文字。若句子读成 “when [linked title] is penetrated…” 或其他不自然结构，应修复周边句法，同时保留 href。

6. DOM 级 HTML 结构保留
Prompt 6 与 Prompt 9 必须做 DOM 级结构检查，不只是视觉或文本检查。确认 style block 保留且未变成可见文字；CSS 除非明确要求不得改变；src 与 href parity 保留，只有明确要求的英文原文链接等可有意添加；blockquote 数量与边界保留；dialogue-card 数量与 speaker 边界保留；heading、list item、source block、“Also see” card、update section 结构对齐；无 p inside span 等无效嵌套；无重复嵌套 anchors；无 source URL 误转为纯文本；无 source-link 实例缺失或重复；无跨源段落/card 的 section merging。

若修复了材料性 HTML 结构问题，状态必须保持 “REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required”，直到后续无材料改动 pass。

7. 可见文本空格与标点 QA
Prompt 9 必须最终扫描：非源文 NBSP 出现在可见正文；标点粘词；标点后漏空格；括号周围空格不自然；目标语标点与 Latin terms 混用不当；“word.word” 或 “واقعهم.ينبغي” 等行合并缺陷；链接相邻标点缺陷。按各目标语言标点规范处理。

8. 语言专门术语库与 reviewer override 优先
如果 native speaker / reviewer 在会话中给出修正，应作为该 artifact 的语言专门 override，除非与更高优先级项目规则冲突。必须搜索全 artifact 相关实例，不只修引用句；标题、heading、正文、cards、重复引用都要一致；并记录改了什么。

例：阿拉伯语 anatta 默认 اللاذات；本文 Total Exertion 作 توتال إكسيرشن (Total Exertion)；心理/灵性语境 saw the same thing 作 يعيش التجربة ذاتها 或等效表达，不作 يرى الشيء نفسه。

9. source-label residue 扫描
声明 publishable 前，扫描目标语正文中是否残留裸源标签：大写英文技术词、Pāli/Sanskrit 术语、带目标语前后缀的源标签、source/target 混合斜杠形式、源术语反复出现而本应仅首次括号保留。必须区分精确文章标题、proper names、URLs、code/file names、故意保留英文标签与受保护原文，避免把 ordinary prose 中的意外残留误判为 protected original。

10. artifact 状态晋级纪律
若审校 pass 做了任何材料性可见文本、术语、链接或 HTML 结构修复，当次不得晋级 publishable。使用：REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required.

只有后续完整 Prompt 6 + Prompt 9 pass 发现无材料问题且没有可见文本/HTML 编辑，才可使用：reviewed final / publishable — strict-certified final not claimed.

11. 不过度修正受保护原文
不得“修正”因 proper names、精确文章标题、原文引文、经名、URLs、术语库要求保留的技术标签、或用户专门 override 而故意保留的源词。但即便保留，也必须自然融入目标语句子。

12. Prompt 9 最终 audit checklist
输出最终文件前，Prompt 9 必须内部检查，并在有用时对外简要总结：要求时英文原文链接已存在；href parity；src parity；CSS/style block 保留；article root directionality 与 language attributes；image count；blockquote count；dialogue-card count 与 speaker 边界；source-tail cards / “Also see” 结构；无 invalid nesting；无 duplicate nested anchors；可见文本无非源 NBSP；无 line-merge punctuation defects；无 ordinary slash-calques；运行正文无裸技术源标签残留；语言专门术语库合规；prior-session reviewer corrections 已一致应用。

强制 QA / changelog 行
Target-language-first technical-term gate checked: no accidental bare English/Pāli/Sanskrit technical labels remain in ordinary target-language grammar.
Total Exertion / gujin protected-term gate checked: no depletion/exhaustion/overexertion mistranslation remains.
Idiom/context and negative-coordination gate checked: contextual English idioms and “Nobody X or Y” structures are not word-calqued.
Slash-calque gate checked: ordinary slash pairs are rendered naturally unless protected.
Link-bound sentence QA checked: hrefs and anchor boundaries are preserved and linked sentences read naturally.
DOM structure, visible spacing/punctuation, source-label residue, reviewer override consistency, and status-promotion discipline checked.



BATCH 79 — thought-watching-thought / no-hidden-witness 加固 — 14 June 2026

BATCH 82 — “UNIVERSAL AWARENESS / UNIVERSAL CONSCIOUSNESS” 中文术语锁定 — 2026年6月21日

本 Batch82 补丁适用于完整 AtR 翻译、审校、QA、对抗性审计、formatter、publishing/controller、prompt-manager、打包、Cloudflare、Blogger 与交接流程中所有涉及中文翻译、审校或验证的环节。

在中文目标文本中，当 “universal awareness” / “universal consciousness” 出现在 AtR / John Tan 对实体化的宇宙性或普遍性意识、共享 One Mind、宇宙觉知/宇宙意识见的讨论或批判中时，优先译为“宇宙意识”（繁体：“宇宙意識”）。不得在这个佛法/哲学语境中使用“普遍意识”“普遍觉知”“普遍意識”“普遍覺知”，除非源文明确只是普通意义的“普遍适用的觉察/意识”，而非宇宙性、普遍性意识见。

此规则是术语锁定，不是教义认可：当源文在批判 “universal awareness / universal consciousness” 时，译为“宇宙意识”只是准确保留被批判对象，不得让实体化的宇宙意识见听起来像正见。

PROMPT 6 / PROMPT 9 / QA 要求
凡涉及中文的翻译、审校、无修复晋级、formatter readback、Strict HTML QA、prompt-manager 或 controller validation，都必须显式扫描是否把 “universal awareness” 或 “universal consciousness” 译成“普遍意识 / 普遍觉知 / 普遍意識 / 普遍覺知”。除非该句明确只是普通“普遍适用”意义，否则应修为“宇宙意识 / 宇宙意識”。

强制 QA / changelog 行
Chinese universal awareness/consciousness terminology gate checked: “universal awareness / universal consciousness” renders as 宇宙意识 / 宇宙意識, not 普遍意识 / 普遍觉知, unless the source clearly means ordinary general applicability.


本 Batch79 补丁为通用补丁，适用于所有 AtR / John Tan 翻译、审校、润色、对抗性审计、格式化、Strict QA、无修复晋级、打包、上传控制、prompt-manager 与交接流程。它适用于所有目标语言，不只适用于阿拉伯语、印尼语或中文。它不得削弱 Batch78 非二元身份句法与 informal idiom 加固；Batch77 agent/做者、efforting/造作用力、mental chattering/概念戏论 假朋友加固；Batch76 高风险佛法术语保护与本地教义污染审查；Batch75 材料性/反反复复改写；Batch74 精确 artifact / 反假通过；Batch73 Presence / hybrid residue；Batch71 Witness/sākṣī；Prompt T；Prompt X；Blogger HTML；href/src/code；以及状态纪律规则。

核心教训
在 AtR / John Tan / anatta 的核心段落中，“thought watching thought rather than a watcher watching thought” 不得译成“心见证心”“心观察心”“觉知见证念头”“意识观看念头”等会暗示更微细的 Mind、Witness、Awareness、observer、knower 或 subject 在背后观看念头的说法。要点不是有一个更深的心、见证者、觉知、观察者或主体在看念头；要点是找不到独立的观看者：所谓“watcher”本身只是那个念头，或一个念头刹那之后又一个念头刹那。

THOUGHT-WATCHING-THOUGHT / NO-HIDDEN-WITNESS 门
对于 “thought watching thought rather than a watcher watching thought,” “the watcher is that thought,” “there is no one behind thoughts,” “first, one thought then another thought,” “Always just this, One Thought!” 以及涉及 watcher, observer, witness, knower, thinker, seer, hearer, mind, thought, awareness, consciousness, subject 的相关 AtR / John Tan / anatta 句子，不得译成会暗示以下任何意思的目标语：
- mind witnesses mind
- mind observes mind
- the witness is the mind
- the observer is the mind
- awareness witnesses thoughts
- consciousness watches thoughts
- 念头背后仍有隐藏的 Witness、Mind、Awareness、observer、knower 或 subject

必须保留 thought versus watcher 的对比，而不是改写成 mind versus observer/witness。应保留 no-hidden-witness 的洞见：thought-after-thought、一个念头观察另一个念头、或 thought watching thought。所谓“watcher”只是那个念头本身，不是背景性的心、见证者、觉知、意识或主体。

当目标语中“THOUGHT”和“MIND”容易混同
若目标语言用同一个词或高度重叠的词来表达 thought 与 mind，译者必须消歧。不要保留隐藏主体歧义，而应使用相当于以下的短语：
- one thought-moment observing another thought-moment
- one arising thought observing another thought
- one mental event observing another mental event
- this thought itself is what is mistaken for the watcher
- the so-called watcher is only that very thought
- there is no observer behind thoughts; only one thought, then the next thought

若字面译法仍容易读成隐藏主体，应在译文中加入小型澄清，而不是保留歧义。

语言示例
印尼语：避免 batin menyaksikan batin、pikiran menyaksikan pikiran、pengamat adalah batin、penyaksi adalah pikiran，或当 pikiran 可能被读成 Mind 而非一个念头时的 pengamat adalah pikiran itu sendiri。优先使用 satu momen-pikiran mengamati momen-pikiran lain；satu kemunculan pikiran mengamati kemunculan pikiran lain；yang disebut “pengamat” itu tidak lain hanyalah pikiran yang sedang muncul itu sendiri；bukan ada seorang pengamat di balik pikiran; hanya ada satu pikiran, lalu pikiran berikutnya。

中文：避免会暗示 心 / 觉知 / 意识 作为隐藏主体在见证念头的表述。视自然度优先使用 念头看念头 / 一个念头看另一个念头 / 所谓的观察者只是那个念头本身。

阿拉伯语：避免暗示 العقل / الوعي / الشاهد 是背景见证者的表述。优先使用相当于 فكرة تراقب فكرة أخرى 以及“所谓观看者只是那个念头本身”的说法，而不是“mind witnesses mind”。

PROMPT 6 / PROMPT 9 / QA 反向回译要求
在每一次 Prompt 6、Prompt 9、Strict QA、formatter readback、无修复晋级、打包校验、prompt-manager、发布/controller 或对抗性审校中，都必须显式扫描涉及 watcher、witness、observer、seer、hearer、thinker、knower、mind、thought、awareness、consciousness、subject、no one behind thoughts、one thought then another thought、thought watching thought、watcher is that thought 的核心 anatta 段落。

运行以下反向回译对抗测试：目标语是否能自然回译成 “mind watches mind,” “mind witnesses mind,” “the witness is mind,” “the observer is mind,” “awareness watches thoughts,” “consciousness witnesses thoughts,” 或“念头背后仍有一个微细主体”？若是，必须修复。此问题一律按 material 处理，因为它会反转或削弱 anatta / no-hidden-witness 洞见。

强制 QA / CHANGELOG 确认行
每一个相关 AtR / John Tan / anatta QA pass 都必须包含以下简短确认行：
Thought-watching-thought / no-hidden-witness gate checked: no “mind witnesses mind,” “witness is mind,” “awareness watches thought,” or hidden-subject reading remains.

状态纪律
若发现并修复此类问题，该 artifact 不能在同一 pass 中晋级。使用：“REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required.” 只有后续精确 artifact、无 material edit 的复审 pass 才可使用：“reviewed final / publishable — strict-certified final not claimed.”

BATCH 78 — 通用非二元身份句法与非正式 idiom / “blah” 语域加固 — Batch81 修订 / 澄清 — 16 June 2026

本 Batch78 规则现由 Batch81 加以澄清。它仍是通用规则，适用于所有 AtR / John Tan 翻译、审校、润色、对抗性审计、格式化、Strict QA、无修复晋级、打包、上传控制、prompt-manager 与交接流程。它适用于所有目标语言，不只适用于阿拉伯语或泰米尔语。此澄清不得削弱 Batch80 hero-title 格式修复；Batch79 thought-watching-thought / no-hidden-witness 加固；Batch77 agent/做者、efforting/造作用力、mental chattering/概念戏论 假朋友加固；Batch76 高风险佛法术语保护与本地教义污染审查；Batch75 材料性/反反复复改写；Batch74 精确 artifact / 反假通过；Batch73 Presence / hybrid residue；Batch71 Witness/sākṣī；Prompt T；Prompt X；Blogger HTML；href/src/code；以及状态纪律规则。

核心教训
某些英文非二元公式与非正式 idiom 若按表层句法直译，会造成结构性风险。在 anatta 语境中，“In X, just Y” 是紧凑的观照式公式。它不是容器/地点公式，但 Batch78 也不应被误解为要求每一种语言都改成沉重的定义句，例如 “hearing itself is sounds,” “seeing itself is forms,” 或 “X means Y”。真正的要求更细：在目标语自然时，尽量保留源文 “in hearing / in seeing / in thinking, only ...” 的简洁节奏，同时防止 hearing、seeing、thinking、awareness 或 mind 被读成一个容器、场域、舞台、背景、实体或隐藏主体，仿佛现象在其中出现。同样，像 “blah something nice but deluded” 这类非正式表达，未必要求译成幼稚、滑稽或过度口语的“叽里呱啦/碎碎念/瞎扯”；它常是在贬斥听起来漂亮但仍属迷妄的概念性/灵性话语。翻译必须保留非二元精确性、紧凑的观照节奏与出版语域，而不是机械保留表层句法。

“IN X, JUST Y” 非二元身份句法门
对于 “In thinking, just thoughts,” “In hearing, just sounds,” “In seeing, just forms, shapes and colors” 这类 anatta / 非二元句子，若目标语自然，应保留相当于 “in/with/at hearing, only sounds; in/with/at seeing, only colors/forms; in/with/at thinking, only thoughts” 的结构。不得使用会暗示容器、类别、地点、场域、舞台、背景、实体、隐藏主体，或过程与对象之间有细微间隔的目标语结构，例如“在思维里面有念头”“在听里面有声音”“声音出现在觉知/心中”等。

这是语义门，不是单一语法模板要求。若会改变源文语气、节奏或教义细节，不得过度修正成僵硬的身份定义句，例如 “X is Y,” “X itself is Y,” 或 “X means Y”。优先使用目标语中最自然、相当于“听时/听中，唯有声音”“见时/见中，唯有颜色与形相”“想时/思维时，唯有念头”的表达，同时确保不会暗示内在容器、背景观察者或隐藏主体。只有当目标语确实需要时，才使用较重的身份/不分离表述来阻断误导性的容器/背景读法。

泰米尔语澄清：若 கேட்பதில் எப்போதும் ஒலிகள் மட்டுமே 会读成内在容器，即“在听里面有声音”，应避免。若 கேட்பது என்பது எப்போதும் ஒலிகளே 或 கேட்பதே எப்போதும் ஒலிகளே 听起来像沉重的 “hearing itself is sounds”，而不是源文简洁的 “in hearing, only sounds”，也应避免。优先使用自然、非容器化的处所/时间表达，例如 கேட்கையில் எப்போதும் ஒலிகளே；பார்க்கையில் எப்போதும் நிறங்கள், வடிவங்கள், உருவங்களே；சிந்திக்கையில் எப்போதும் எண்ணங்களே；கேட்கையில் வெறும் ஒலியே。

阿拉伯语专门说明：较早的身份式例子，如 التفكير ليس إلا أفكارًا / السماع ليس إلا أصواتًا / الرؤية ليست إلا صورًا وأشكالًا وألوانًا，在需要阻断地点/容器读法时仍可使用。但若当地句子需要保留源文紧凑的观照节奏，不得机械强迫使用这些重身份句。避免 في التفكير، مجرد أفكار，若它可能读成地点/类别结构而非非容器化的 “only thoughts” 公式。

非正式 idiom / “BLAH” / 贬抑性概念话语门
当源文使用 “blah something nice but deluded” 这类非正式表达时，不得机械译成目标语中相当于“闲扯、唠叨、碎碎念、胡扯、聒噪”的词，若这会造成非源文意图的滑稽、幼稚、过度随便、或只是“多嘴/话多”的语气。

必须保留其功能：贬斥好听但迷妄的概念性话语、漂亮的表述、或缺乏证悟的悦耳灵性语言。所有目标语言都应以适合目标语出版语域的自然表达来处理，例如相当于：“说出听起来漂亮但仍然迷妄的话”“编织出响亮/漂亮的句子，但仍属迷妄”“说出动人的套语，但仍然迷妄”“给出优雅好听的概念性说法，但仍未离迷妄”。

阿拉伯语专门规则：优先使用 صاغوا عبارات رنّانة جميلة، لكنها واهمة / أطلقوا كلامًا منمقًا جميلًا لكنه واهم。除非源文明确是在非常口语化地嘲讽，否则避免 ثرثروا。

PROMPT 6 / PROMPT 9 / 对抗性审校要求
在 Prompt 6、Prompt 9、Strict QA、formatter readback、打包校验、无修复晋级、prompt-manager/controller 审查或任何对抗性审计/审校中，必须显式扫描每一种目标语言中的 Batch78 身份句法门的两面：
1. 保护不足：“In X, just Y” 被译成地点、容器、类别、场域、背景、隐藏主体，或“在/里面/其中有……”结构，从而削弱非二元不分离；
2. 过度修正：把源文紧凑公式强行改成 “X is Y,” “X itself is Y,” 或 “X means Y” 之类沉重定义句，导致语气、节奏或教义改变；
3. “blah” 等非正式英文 idiom 被过度字面化为 chatter/prattle/babble 式说法，而没有保留其贬抑功能与适当出版语域。
若任一问题改变源文意义、教义清晰度、非二元精确性、紧凑观照节奏或出版语域，即为材料性问题。必须全局修复，并视情况更新 Source-Term Decision Table / 高风险术语 ledger / Change Materiality Ledger，并使用：“REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required.” 只有后续单独一次精确 artifact、无材料编辑的重审，才可使用：“reviewed final / publishable — strict-certified final not claimed.”

强制 QA / changelog 确认行
每一次相关的 AtR / John Tan / anatta QA pass 都必须包含以下简短确认行：
Batch78 identity-syntax gate checked: no container/background/hidden-subject reading remains, and no over-heavy “X is Y” identity overcorrection was introduced.

BATCH 77 — 通用 agent/做者、efforting/造作用力、mental chattering/概念戏论 假朋友加固 — 13 June 2026

本 Batch77 补丁为通用补丁，适用于所有 AtR / John Tan 翻译、审校、润色、格式化、QA、无修复晋级、打包、上传控制与交接流程。它不得削弱 Batch76 高风险佛法术语保护、本地教义污染审查、反实体化/反虚无化规则，Batch75 材料性/反反复复改写规则，Batch74 精确 artifact / 反假通过规则，Batch73 Presence / hybrid residue 规则，Batch71 Witness/sākṣī 规则，以及 Prompt T、Prompt X、Blogger HTML、href/src/code 与状态纪律规则。

核心教训
在 AtR / John Tan / anatta 翻译中，普通字典义可能成为假朋友。“agent” 在此类语境中常指被预设为行动来源的独立做者、行动者、控制者或主宰者，不是中介、代理人、代表或服务代理。“efforting” 常指造作的、自觉用力的、构造出来的 subtle doing，并不等于疲劳、费力到累、艰苦或耗神。“mental chattering” 在技术性佛法/哲学说明中常指概念化、推论化、增殖性的言说性思维，而不只是口语化的“脑内碎碎念”。只要上下文涉及 anatta、no-self、no-doer、dependent origination、non-doership、修行、知见、证悟或 AtR 说明，就必须把这些词作为高风险术语处理。

AGENT / DOER / AGENCY / DOERSHIP / AGENTLESSNESS / NON-DOERSHIP 审查门
在 anatta、no-self、no-doer、no-doership、agentlessness、dependent origination 及相关佛法/哲学语境中，“agent” 指被预设为独立做者、行动者、控制者、第一推动者或行动所从出的实体。除非源文清楚是在普通非佛法意义上使用，否则不得译成中介、经纪、代表、旅游代理、法律代理、商业代理、软件 agent、服务 agent 或被委托的代理人。

所有语言都必须选用能保留“独立做者/行动者/控制者”义的目标语表达。若目标语言中普通的“agent”对应词会导向中介/代表义，必须拒绝，并改用“independent agent/doer（独立 agent/做者）”、“independent actor/controller（独立行动者/控制者）”、“presumed doer（被预设的做者）”或相应目标语释义。

阿拉伯语专门规则：优先使用 فاعل، فاعل مستقل، الفاعل، الخلوّ من الفاعل، غياب الفاعل المستقل。避免 واسطة، وسيط، وكيل، عميل，以及任何暗示中介、代表、经纪、客户、商业/法律代理或服务代理的词，除非源文清楚是在该普通意义上使用。

EFFORTING / EFFORTLESSNESS 审查门
在 John Tan / AtR 用法中，“efforting” 常指造作、构造、自觉用力、人为努力或 subtle doing，是自然无造作/effortlessness 的反面。它通常不是疲惫、劳累、艰辛、吃力、费劲或身心疲劳。

所有语言都必须避免把 “efforting” 译成“累人的、艰苦的、令人疲惫的、费力到疲劳”等，除非具体句子清楚是在说疲劳。优先使用相当于“造作的努力”“人为努力”“构造出来的 doing”“带有造作的用力”“自觉用力/刻意 efforting”的表达。

阿拉伯语专门规则：优先使用 التكلّف، جهد مصطنع، مصحوب بالتكلّف، ينطوي على جهدٍ مصطنع، متكلّف。避免 مُجهد، مُرهق، شاق，除非源文清楚是在说疲劳、耗竭或身心艰苦。

MENTAL CHATTERING / DISCURSIVE THOUGHT 审查门
当 “mental chattering” 出现在 AtR / Dharma 说明中，并作为对概念性、推论性、增殖性、言说性/戏论性思维的技术性批评时，不得在目标语中机械译成过度口语化的“mind chatter”式表达，若该译法会降低文体、让哲学/佛法语境变得轻浮或随便。应保留“discursive/conceptual proliferation（言说性/概念性增殖、戏论、概念活动）”的意思，并选择适合该语言的佛法/哲学语域。

阿拉伯语专门规则：优先使用 التفكير الاستطرادي، التفكير الخطابي/المفهومي, حركة الذهن الخطابية/المفهومية。只有在源文语气明显口语或非正式时，才可使用 الثرثرة الذهنية。

PROMPT 6 / PROMPT 9 / 对抗性审校要求
在 Prompt 6、Prompt 9、Strict QA、formatter readback、打包校验、无修复晋级或任何对抗性审校中，必须显式扫描每一种目标语言中的以下假朋友：
1. “agent” 被误译为中介/代表/经纪/代理/服务代理，而不是独立做者/行动者/控制者；
2. “efforting” 被误译为疲劳、劳累、艰辛、费劲或耗竭，而不是造作/自觉用力的 doing；
3. “mental chattering” 在技术性佛法/哲学语境中被译得过度口语化，损害文体或意义。
若这类问题改变源文意义、教义清晰度或出版语域，即为材料性问题。必须全局修复，更新 Source-Term Decision Table / 高风险术语 ledger，并使用：“REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required.” 只有后续单独一次精确 artifact、无材料编辑的重审，才可使用：“reviewed final / publishable — strict-certified final not claimed.”

BATCH 76 — 通用高风险佛法术语保留 + 本地教义污染审查 + 反实体化/反虚无化强化 — 8 June 2026

本 Batch76 补丁为通用补丁，适用于所有 AtR 翻译、审校、润色、格式化、QA、无修复晋级、打包、上传控制与交接流程。它不得削弱 Batch75 的材料性润色/反反复复改写规则、Batch74 的精确 artifact / 反假通过规则、Batch73 的 Presence / hybrid residue 规则、Batch72 的 Plum Dawn / 复制按钮 / UI 规则，以及更早的 Blogger chrome、href/src、不增添解释、不受公开译本污染、Prompt T、Prompt X 等规则。

核心教训
佛法技术术语在目标语言中可以语法正确、字典有据、网上有人使用，但仍可能在教义上不安全。某些本地词汇可能带入造物主/本源论、苏菲/伊斯兰、基督教、吠檀多/印度教、道教、新时代、心理学、法律、民间宗教、虚无论、常见论、实体化或形而上学色彩。因此，高风险佛法术语不得自动采用“最自然”的本地等值词。

通用 Source-Term Decision Gate（源术语决策门）
在初译定稿前，以及 Prompt 6 / Prompt 9 / Strict QA 式审校中，必须为文章核心高风险术语维护 Source-Term Decision Table。每个术语至少分类：
1. 源术语。
2. 拟用目标语译法。
3. 风险类型：安全可译 / 保留原术语 / 保留原术语 + 目标语释义 / 目标语词仅可在澄清后使用 / 因教义污染而拒绝该目标语词。
4. 理由。
5. 本 artifact 的最终处理规则。
若没有已知安全译法，应保留 English/Sanskrit/Pāli/Tibetan/AtR 标签并加简明目标语说明，而不是自信地创造一个本地等值词。

高风险术语族
至少对以下术语启用此门：dependent origination / dependent arising / pratītyasamutpāda / paṭicca-samuppāda；emptiness / śūnyatā；anatta / anātman / no-self / non-self / not-self；Self/self/true Self/self-view；I AM / AMness / Isness；Presence；Awareness/consciousness/mind；rigpa / vidyā / luminosity / clear light；Madhyamaka / Middle Way；Buddha-nature / tathāgatagarbha；non-arising / unborn；non-reification；suchness / thusness / dharmadhātu / dharmatā；total exertion / dependent designation / twofold emptiness；witness / Witness / sākṣī；agency/agent/actor/doer/perceiver；fruition/Buddhahood/liberation/awakening/enlightenment；以及文章特有的高风险术语。

本地宗教污染审查门
除非源文明确作比较，不得把佛法术语暗中翻成本地宗教/神学词汇。不得作为默认隐性译法的例子包括：苏菲/伊斯兰词如 fanāʾ、baqāʾ、ihsan、tawḥīd、Allah-realization；基督教词如 salvation、grace、Holy Spirit、Godhead、soul、creator；吠檀多/印度教词如 Brahman、Ātman、mokṣa、īśvara；道教词如 Dao、wu wei、original qi；新时代/心理学词如 universal consciousness、higher self、subconscious、ego death、pure consciousness。这些词只有在源文本身明确比较时，才可作为显式比较说明出现，不能成为佛法术语的隐藏译法。

反实体化 / 反本源论审查门
对于 dependent origination、emptiness、Buddha-nature、Presence、Awareness、rigpa、dharmadhātu、ground、source、essence、luminosity 等相关术语，检查目标语是否暗示固定本源、起源实体、独立根基、造物主、Brahman/Ātman、永恒见证者、常住 Self、形而上容器、万法从自存基底中流出、源文不含的对称互依、或把广义缘起/依他施设降格为单纯线性因果。

反虚无 / 反空洞审查门
对于 emptiness、voidness、nothingness、absence、non-arising、no-self、cessation 等相关表达，检查目标语是否暗示一切都不存在、断灭、无意识空白、死寂空洞、单纯缺席、抑郁/无意义、存在论非有，或把空性当作实体/地点。若本地词有这种风险，应保留 śūnyatā/emptiness 或相关源标签，并说明：空无自性/自有，不是虚无。

受保护标签与转写规则
当 AtR 或源文惯例需要时，必须精确保留受保护标签，尤其是 I AM、AMness、Isness、大写 Self、技术性 Presence、rigpa、vidyā、必要时的 anatta/anātman、Madhyamaka、śūnyatā，以及目标语等值词有风险时的 dependent origination。可以加目标语释义，但不得静默替换受保护标签。除非某英文技术术语的音译在该语言中已经是稳定惯例且适合本文，否则不要把英文术语音译成目标文字；优先保留拉丁源标签并加目标语说明。

RTL / 混合文字安全
对于阿拉伯语、希伯来语、波斯语、乌尔都语等从右到左书写页面，插入 Latin/Sanskrit/Pāli/Tibetan 技术标签时，必要时必须使用 bidi-safe span 或同等方向隔离。Formatter 与 Strict HTML QA 必须检查混合文字标签是否可读、未被重排、断裂或搅乱。不得把 bidi-safe spans、术语说明、首次出现释义当作“冗余”清理掉。

专家反馈门
若母语者或具领域知识的审校者指出某目标语术语造成教义扭曲，必须分类为：材料性教义风险 / 目标语流畅度问题 / 可选偏好 / 拒绝建议。不得仅因该词出现在字典、Wikipedia、旧译、或网上用例中就忽略。若属材料性问题，须全局修复，更新 Source-Term Decision Table，并写入 Change Materiality Ledger。

各提示词应用
Universal Intro 与 Prompt A：显式允许用户/审校者提供专家术语警告；要求 Source-Term Decision Table、首次出现处理与有意添加释义的记录。
Prompt 1/2/3/4/5/7/8：不要过度本地化高风险佛法术语；当本地等值词不安全时，保留标签并加释义；Prompt 8 不得把对话中的技术术语润色成本地宗教等值词。
Prompt 6：加入 Local Doctrinal Contamination / Anti-Substantialism / Anti-Nihilism 对抗性审查篮，并全局扫描高风险术语族。
Prompt 9：自然流畅不得压过术语安全；不得把受保护标签替换成听起来自然但教义污染的本地词。
Prompt T 与 Prompt X：当源锚点/源文恢复流程会进入目标 artifact 时，加入反本地宗教污染、受保护标签保留、释义一致性检查。
Protocol A/B 与 Upload Controller：要求 Source-Term Decision Table、术语保护说明、有意释义、专家反馈纳入、材料性术语修复后的状态纪律。
Unified Blogger Formatter 与 Strict HTML QA：保留术语说明、受保护标签、释义、bidi-safe spans、混合文字标签；审查受保护技术标签、首次出现释义、本地宗教污染、本源论/实体化、虚无化与方向性安全。
RemoveSegID / 机械工具：不得改动受保护技术标签、术语说明、释义、bidi spans、源标签、href/src/code 或解释性术语说明。

阿拉伯语案例示例——只是例子，不是阿拉伯语专属规则
dependent origination 不应在审校者警告其可能含有本源论时，只依赖 النشوء المعتمد / النشأة المعتمدة 等阿拉伯词。更安全的高风险处理可保留 “dependent origination (pratītyasamutpāda / paṭicca-samuppāda)” 并加阿语说明：قيام الظواهر اعتمادًا على شروط وأسباب متداخلة، لا صدورها من أصل ثابت أو مصدر قائم بذاته。
emptiness / śūnyatā 若用 الفراغ 会暗示空洞或虚无，应改用释义：الخلوّ من الوجود الذاتي أو الطبيعة الذاتية، لا العدم ولا الفراغ بمعنى الانمحاء。
I AM 作为 AtR 阶段标签保留 I AM，不应变成苏菲词。fanāʾ、baqāʾ、ihsan、tawḥīd 等苏菲/伊斯兰词，除非源文明确比较，否则不能作为佛法术语主译。Madhyamaka 在宗派名称身份重要时，不应只化约为本地语言的 “Middle Way”。

材料性与状态
凡涉及 dependent origination、emptiness、anatta、Self、I AM、Presence、Awareness、rigpa、Madhyamaka、Buddha-nature、agency/doer/perceiver 或其他高风险术语的术语改动，均属材料性改动。此类 pass 不得在同一轮晋级为最终。状态必须是：“REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required.” 只有之后重新打开精确 artifact、无材料编辑的 readback pass，才可使用：“reviewed final / publishable — strict-certified final not claimed.”


适用范围： Prompt T: 应用 to Indic/Tibetan source-anchor decisions when Presence, rigpa/vidyā, anatta/anātman, śūnyatā, total exertion analogues, or protected labels overlap.




第 75 批 — 普遍适用的“改动重要性判断 + 防反复打磨”强化 — 4 June 2026

本第 75 批补丁普遍适用于所有 AtR 翻译、审校、润色、修复、格式化、QA、无修复晋级、打包与交接流程。它不会削弱第 74 批关于防假通过、精确工件、拉丁字母语言流畅度、混合残留、标题/链接、href/src、Blogger 原生结构、源文忠实度等任何硬性关卡。

改动重要性判断关卡（Materiality-Aware Polish Gate）
在任何翻译、润色、对抗性审查、修复、审计或无修复晋级中，只要准备改动可见文本，必须先分类：

A. 必须修复（Required repair）——必须改。
包括影响源文忠实、遗漏/添加、误译、教义或术语准确性、核心术语一致性、说话人归属、精确标题/链接文字、源语言残留、混合语言/混合标签残留、保留的梵文/巴利文/藏文/英文标签形态错误、目标语言语法、标点或断行合并损伤、可读性、HTML/链接/媒体/代码完整性，以及目标语表达已经显得机器翻译、生硬、晦涩、干扰阅读或误导教义理解的情况。

B. 有益润色（Beneficial polish）——主动编辑阶段允许。
当前译文不一定严格错误，但改动能实质提升发表质量：更自然、更顺、更符合公开 AtR 佛法文章语体、减少翻译腔、澄清难懂的教义点、统一重复术语家族、把生硬直译改成自然表达、让保留技术标签更自然地融入目标语语法。若改了可见文本，该轮仍属于编辑/修复/润色轮，除非之后另开无材料改动的晋级轮，否则不能在同一轮宣称无修复晋级。

C. 偏好性改写（Preferential rewrite）——避免。
如果现有表述已经忠实、教义清楚、语法正确、自然、术语一致、可发表，并且没有残留、混合标签、标题漂移、标点、链接或 HTML 问题，不要仅因另一个同义表达也可以或稍微偏好就反复改写。

模式规则
主动翻译/修复/对抗性审查：必须修复 A 类；可以进行 B 类发表质量润色；通常避免 C 类偏好改写。凡改动可见文本，必须记录并保持修复/编辑状态。

无修复晋级：必须重新打开精确最新版工件，主动寻找必须修复的问题，也可识别有益润色但不得编辑，不能只因还有另一个可接受说法就阻止晋级。若只剩可选润色或偏好替代，记录“仅有可选润色；无需材料修复”。若发现必须修复问题，则修复并重置为修复状态。

改动重要性台账
修复或对抗性审查报告必须包含简明“改动重要性台账”：最近标题/定位点、旧文、新文、类别（必须修复 / 有益润色 / 已避免偏好改写）、理由、是否影响状态。无修复晋级报告必须列明：是否发现必须修复、是否发现但未编辑有益润色、是否避免偏好改写、是否做了材料改动。

防反复打磨停止规则
多轮对抗性审查之后，如果精确最新版工件没有必须修复的问题，剩余只是偏好改写或可选有益润色，就停止编辑。不得因同义词级别或偏好级别改写反复重置状态。只有在重新打开精确工件、比对源文、通过术语/残留/混合/标题/链接/href/src/结构/流畅度关卡、没有必须修复、没有可见文本材料改动后，才能进入无修复晋级。

不得削弱强 QA
遗漏、添加、误译、教义扭曲、错误术语、源语言残留、混合语言残留、保留标签形态错误、语法损坏、影响发表质量的翻译腔、精确标题漂移、链接文字损坏、href/src/媒体/代码突变、说话人归属错误、标点或断行合并损伤、结构遗漏/重复、误导性的目标语表达，仍然都是必须修复的材料缺陷。

BATCH 74 — UNIVERSAL MATERIALITY-AWARE POLISH + ANTI-CHURN HARDENING — 2026 年 6 月 3 日

本 Batch76 补丁是通用补丁，不是只针对波兰文。它适用于拉丁字母、非拉丁字母、屈折语、分析语、从右到左书写语言，以及混合文字目标语。必须保留 Batch73 Presence / hybrid-residue 规则，以及 Batch72/71/70/68/67/66 继承下来的所有防护。

虚假通过 / 重用旧 artifact 失败门
任何 review、repair、no-repair promotion、package 或 handoff pass，只有在重新打开 exact latest artifact 并为本次 pass 产生新的证据时才有效。不得把过去的 pass summary、旧链接、旧 QA 报告、旧文件名或记忆当作新的 review。若助手发现自己声称做了新 pass，但实际只是重用旧 artifact 或没有真正 review，必须明确撤回该说法，指出最新真实 artifact，不得继续假装该 false pass 存在。

Exact-artifact 证据要求
每次 pass 都必须报告 exact input artifact path/name、exact output artifact path/name、输出是否 byte-for-byte identical 或 materially changed，并在工具环境允许时给出 SHA/hash 或等效 identity evidence。还必须说明每个 intended replacement 是否真的匹配并应用。若没有生成新 artifact 或新的 no-material-edit QA report，不得声称完成新 pass；应说：No new pass was completed.

No-repair 与 repair 必须分离
严格 no-repair promotion pass 不得做 material edits。若发现任何 material issue，promotion 失败。要么停止并输出 failed-promotion report，要么另开 material repair pass。任何做了 material edits 的 pass 都必须使用：REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required。不得在同一 pass 中同时 promotion。

禁止 thought-only review
只思考几秒后重复旧链接不是 review。有效 review 必须有具体证据：打开 latest artifact、搜索/扫描目标文本、抽查高风险 source/target passages、运行目标语 fluency checks、检查 cumulative risk basket、检查 href/src/media/code parity，并生成新的 QA report，或明确说明没有创建新 artifact。

字符串式修补脆弱性 / 0-match 规则
如果 targeted repair 在当前 artifact 中没有匹配，不得假设问题已经修好。必须搜索语义变体、较短子串、屈折变体、大小写变体、标点变体、HTML 分隔变体，以及附近上下文。每个 0-replacement intended repair 都要记录为 warning 并要求调查。脚本修补的 QA 必须列出搜索的 old string/pattern、match count、replacement applied、0-match intended repairs、semantic follow-up search，以及 cumulative risk basket 的前后状态。

拉丁字母目标语 residue 与 fluency 门
拉丁字母目标语中，普通英文 residue 更难看出，因为源语与目标语都可能使用拉丁字母。必须额外运行 Latin-script residue scan，检查普通 prose 中残留英文、英文技术标签造成当地语法错误、英文术语与当地屈折/格/介词粘连、未加引号的英文标签被当作目标语名词、source title fragments 被部分翻译，以及看似通顺但实际是英文成语直译的 calque。

屈折语言 protected-label agreement 门
对于屈折目标语，保留的 AMness、I AMness、Presence、Awareness、Self、No Mind、no-mind、Dharma、Dhamma、Maha、anatta、anātman、dhātu、skandha、śūnyatā 等标签，必须检查格、性、数、agreement 与自然语法。若保留 protected label，周围句子仍必须自然。必要时改用目标语在前、引号、或短说明，而不是强迫 retained label 承受目标语语法。

Slash-calque hardening
扫描可见 prose 中的 slash compounds，例如 A/B、A / B、A/B/C、subject/object、doer/agency、Dhamma/Dharma、I/self/Self、mind/consciousness、realization/actualization、no-self/non-self/anatta/anātman、链接标题片段，以及用 slash 连接的人名/别名。每处都必须分类为 exact title/proper label、glossary/termbank、ordinary prose 或 ambiguous。精确标题与有意 glossary label 可保留；ordinary prose 应改成自然的目标语并列、同位语或说明；ambiguous 则标注 human review。

精确标题 vs 翻译标题强化
每个 link-bound title 都要单独决定：完整保留精确原题、完整自然翻译，或有意配对。不得部分翻译标题、尊称或标题片段，例如把 “Mr.” 译成目标语却保留其余英文。不得把目标语词语混入精确英文标题，除非原题本来如此。QA 必须列出 linked titles 是 preserved、translated，还是 intentionally mixed/paired。

Idiom / metaphor calque hardening
Prompt-level review 必须主动搜寻英文成语/比喻的机械直译，并建立目标语专属 idiom/calque risk basket。种子例子包括：go and read、goes down the drain、along the line of、break through、stood out、one hand claps、the drop is thorough、quality time、open and boundless、wholly thus、always already so、no mirror reflecting、magical appearances are too empty。本列表只是示例，不是穷尽。

技术术语族漂移门
一旦某术语族造成反复修补，就必须加入 cumulative risk basket 并全局扫描所有变体。高风险术语族包括 actualization/actualize/embodied/lived/integrated；ground/Ground of Being/ground reality/basis/path；Presence/pure Presence/total Presence；AMness/I AMness/I AM；Self/self/no-self/no-I/anatta/anātman；no-mind/No Mind/Mind that is no-mind；subject/object；doer/agency/no-doership；total exertion/total activity/full functioning；suchness/thusness/isness/wholly thus/Always So；Maha；yāna/lower yānas；dhātu/skandha；knowledge/cognitive obscurations；māyā/magical appearances；mirror/no mirror/mirror-like awareness。

机械正确但不自然也是 material
译文即使通过术语表，也可能因目标语 prose 不自然或教义上误导而失败。若影响可读性或教义清晰度，以下都属于 material defects：格/性/数 agreement 不自然、源语词序复制太重、line-merge 造成片段句、技术标签未整合进语法、slash-heavy prose、比喻直译、链接/引号/括号附近目标语标点问题、技术词大小写问题、把 Dharma 技术词误译成普通词。

No-repair promotion 证据附录
No-repair pass 必须包含 “No New Artifact Reuse / Exact Artifact Evidence” 小节，列明：I reopened: [exact file]；I produced: [exact report/file]；Material edits made: yes/no；若 no，输出是 byte-for-byte identical / hash-equivalent / explicitly copied unchanged；若 yes，则 no promotion allowed；未假设 prior pass status；列出 new review evidence。

跨语言术语来源证据的限度
在可行时，可用目标语佛教资料确认主要 Dharma 术语选择，但这类证据只支持大方向术语选择，不证明句子级忠实度、上下文适配或目标语流畅度。段落/链接/blockquote/list 数量一致是必要但不足的，绝不能当作翻译质量证明。

状态纪律
若在先前 no-repair promotion 后又发现 material issues，必须写明：Previous promotion was premature. This pass found material issues that should have been caught earlier. Status is reset to REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required. 永远不得声称 strict-certified final。

BATCH 73 — UNIVERSAL PRESENCE GATE + TARGET-LANGUAGE-FIRST PARENTHETICAL ORDER + HYBRID RESIDUE HARDENING — 1 June 2026

本 Batch73 补丁是通用补丁，不是只针对日文。凡是会创建、翻译、审校、精修、HTML 翻译、审计、格式化、打包、发布或声明 ATR artifact 状态的 managed prompts，均须应用本规则。

通用关键术语研究门控
翻译或统一任何高风险 AtR / 佛法术语前，不得只依赖字典、记忆、既有目标语习惯，或另一种语言的术语表。每篇文章应为文章核心术语建立或更新简明的 Key Term Research / Decision Table。表中至少记录：源词、目标语译法、翻译/保留/转写/配对状态、是否可作为 protected label 保留、普通 prose 译法规则、禁用或高风险译法、是否需要外部/源文研究、是否有多重语义需按上下文区分。

最低高风险篮：Presence/presence、Awareness/awareness、I AM/AMness、anatta/no-self/non-self/anatman、emptiness/empty nature、dependent arising/dependent origination、non-dual/non-duality、no-mind/No Mind/One Mind、luminosity/clarity/vividness、manifestation/appearance/phenomena、suchness/thusness/isness、subject/object/subject-object division、agent/agency/doer/doership/non-doership、spontaneous perfection、total exertion、Dzogchen/Mahamudra/rigpa/Dharma/Buddhadharma，以及文章特有核心术语。

通用 Presence Gate
Presence 必须视为高风险 AtR / 佛法术语，不可当成普通英文 presence 机械处理。必须区分技术性 AtR Presence、普通 presence/present、ever-present / already-so 句式、西方哲学 / metaphysics of presence 语境，以及 Dzogchen 标题或固定短语如 Instant Presence。技术性 AtR Presence 指 luminous immediacy / direct actuality / vivid fact of being-experience，常见于 pure Presence、Total Presence、Presence-Awareness、Presence is spontaneously perfected、manifold of Presence、nature of Presence 等表达，须经目标语研究和术语表决定。

日文 Seven Stages 文章族的 house-style 说明：技术性 AtR Presence 默认用 臨在；pure Presence = 純粋な臨在；Total Presence = 総体的臨在；Presence-Awareness = 臨在＝覚知；Presence is spontaneously perfected = 臨在は自ずから円成している / 按上下文处理。除非上下文明确是哲学/现象学语境，不得把 現前性 作为 AtR Presence 默认译法。普通 present / ever-present 可按句子需要使用 現前している、すでに現れている、常にすでにそうである 等自然表达。此日文说明不得强加给其他目标语。中文 AtR house style 可在适当处保留 臨在，但中文选择只是线索，不自动决定所有其他语言。

目标语优先括注顺序规则
在普通目标语 prose 中，如需保留英文/Sanskrit/Pāli/Tibetan/source label，默认使用目标语在前：目标语译法（source label）。普通 prose 中避免 source label（目标语译法），除非该 source label 是精确标题、正在作为短语被讨论、目标语惯例强烈要求 source label 在前、用户明确要求，或该处是 glossary/termbank。不得把保留的 source label 直接粘到目标语格尾、助词、后缀、后置词、clitic 或屈折结尾上。

日文例子：用 無我（anatta）の，不用 anatta の；用 空性（shunyata）の実現，不用 shunyata の実現；普通 prose 用 心（Mind），不用 Mind（心）；用 伝達（transmission），不用 transmission（伝達）；用 無我（no-self），不用 no self（無我）。避免 Pure覚知、Natural覚知、Non-dual は、dependent origination の、ground と path、meditative composure と insight、no-mind の状態、anatta実現、Maha経験、On 空性、Self-実現、Being／Self-Realisation への 等 hybrid。

通用 hybrid residue scan
最终稿或 promotion 前，必须建立目标语专用 hybrid scan basket。扫描：英文/source label 粘连本地助词/格标记/后缀/屈折；英文词进入目标语复合词；目标语词插入精确英文标题；mixed-script 技术 hybrid；普通 prose 中英文优先括注；精确英文标题中夹入已翻译片段；普通英文 prose 被误保护成 technical label；括号后本地助词前出现异常空格。拉丁字母目标语须额外做 ordinary-English residue scan，因为英文残留更难发现。

日文风险篮种子：Pure覚知；Natural覚知；Non-dual は；dependent origination の；ground と path；meditative composure と insight；no-mind の状態；anatta の/へ/と/を/は/後/実現；Maha経験；On 空性；Self-実現；Being／Self-Realisation への；transmission（伝達）；no self（無我）；Dhamma／Dharma） を；Awakening to Reality 実践 Guide；音Cloud；Scenery……；真如のMaha経験；五蘊（五蘊）。并扫描 Latin/source label + の、を、に、へ、と、は、が、で、も、から、まで、より，以及 後、的、性、実現、経験、状態、化、作用、実践、知見、覚知、空性。

精确标题 vs 翻译标题处理
每个带链接的文章/书/video/audio 标题都必须判断：link text 应保留精确 source title，还是译成完整自然目标语标题。若保留，则整题精确保留，不可局部翻译；若翻译，则整题自然翻译，必要时在括号中保留 source title。不得制造 On 空性、Awakening to Reality 実践 Guide、Recognizing Rigpa vs Realizing 空性 等混合标题。不得翻译 URL slug。QA 必须记录有意保留或有意翻译的标题。

Protected-label whitelist 纪律
某词是 technical 不代表它必须留英文。每个保留的 English/Sanskrit/Pāli/Tibetan label 必须归类：精确标题、专名、URL/code、有意保留的技术术语、有意保留的 English AtR label、目标语译法后的括注 gloss、或 human-review placeholder。ordinary lowercase terms 如 experience、realization、practice、view、awareness、consciousness、manifestation、ground、path、source、self、mind、state、clarity、naturalness、ordinariness、spontaneity、agent、doer、doership、subject、object 通常应翻译，除非记录保留理由。

术语族升级规则
若同一术语族出现一次以上 material issue，必须把该术语族升级到 active termbank 和 scan basket。例如 Presence 曾误译一次，则全局扫描 Presence、pure Presence、Total Presence、Presence-Awareness、nature of Presence、present、ever-present 及目标语候选；若出现一个 anatta hybrid，则扫描所有 anatta/no-self/non-self/anatman；若出现 total exertion 问题，则全局扫描 total exertion/total activity/complete activity/full function 等等。

Total exertion 源文觉察规则
不得机械给 total exertion 设默认译法。日文 Dōgen/Zenki 语境应按源义考虑 全機 与 一法究尽。中文继续保留既有规则：不得把 total exertion 默认译成 全体作用，应使用 一法究尽 / 具尽 / 依上下文处理。其他语言须先研究源文/语境，再定默认。

No-repair promotion 强化
No-repair promotion 不得只做 count、href/src 和结构检查。必须重新打开 exact artifact、对照 source、运行 Prompt 6 语义忠实审校、Prompt 9 目标语流畅度审校、目标语 hybrid residue scan、目标语优先括注顺序扫描、高风险术语表一致性扫描、精确链接标题扫描、普通 prose vs protected label 分类、URL/href/src/code 完整性扫描、标点/并行/断行扫描，并从开头、中段、结尾、blockquote、粗斜体区和所有文章核心术语处抽样审读。若此前已经 promotion 后又发现 material issue，必须明确撤回先前 promotion 并重置状态：Previous promotion was premature. This pass found material issues that should have been caught earlier. Status is reset to REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required.

Required QA evidence additions
每份 final/review/promotion QA report 应包含：确切 source file、确切 target file、使用的 prompt versions、目标语风险画像、key term decision table、Presence Gate 结果、括注顺序扫描、hybrid residue scan basket、ordinary English/source residue basket、protected-label whitelist、精确链接标题扫描、href/src/media parity、URL/code 中是否混入非源文本、主要结构计数、代表性句子抽样、累计风险篮与结果、是否做了 material edits、正确状态标签。

状态纪律
Batch73 属 material managed-prompt update，状态必须为：REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required。只有后续 exact-artifact no-material-edit readback pass 才可提升为：reviewed final / publishable — strict-certified final not claimed。

BATCH 71 — UNIVERSAL WITNESS/SĀKṢĪ GATE + RESIDUE/CALQUE HARDENING — 2026 年 6 月 1 日

本 Batch71 补丁源自 Tamil Seven Stages 审校经验，适用于所有会创建、翻译、编辑、审校、精修、审计、格式化、打包、发布或声明 translated ATR artifacts 状态的 managed prompts。保留 Batch70 hard no-repair promotion discipline、Batch68 非拉丁残留规则、Batch67 dialogue/source-box parity、Blogger chrome preservation、exact href/src parity、no-added-gloss、no-public-translation-contamination、Prompt T source hierarchy、Prompt X source-restoration rules 与既有 Soh/AtR 中文术语表。不得更新独立维护的 standalone user intro prompt，除非用户明确要求。

UNIVERSAL WITNESS / SĀKṢĪ GATE
English witness / Witness / Eternal Witness / Witness-consciousness / watcher / observer / seer / hearer / perceiver / knower / subject 属高风险词，不得机械翻译。翻译或批准前必须分类：A. 普通法律/社会见证者；B. Advaita/Vedanta/Self/Witness-consciousness 语境，应使用目标语中表示 Sākṣī / sākṣin / Witness-consciousness 的哲学技术表达，不可只用法律式“证人/见证人”；C. Buddhist no-self/anatta critique 语境，必须保留被否定的是 Advaita-style ultimate Witness/Sākṣī/background subject，而非法律证人或普通观察者；D. Indic 或受 Indic 影响的目标语可用本地 Sanskritic form，例如 Tamil சாக்ஷி、Hindi साक्षी、Bengali সাক্ষী；E. 非 Indic 语言若普通 witness 词太法律化，应使用 Sākṣī、witness-consciousness、Witness-Self、pure witnessing awareness 或 witnessing subject 等技术表达。QA/changelog 必须说明 Witness/Sākṣī 如何分类与处理；未决则列入 human review。

VISIBLE LINK TEXT 分类门控
可见 <a> 文字仍是文章可见文本。不得自动保留所有英文链接文字。每个链接文字节点必须分类为 URL-only display text、proper name/person/group/source title、exact article/book/video/audio title、sentence-like prose containing a link、或 hybrid title/prose fragment。句子式链接 prose 应自然翻译，href 必须 exact-preserve。把普通英文句子误当 protected title、翻译/破坏 href、或让目标语文章中残留普通英文 link prose，均属 material defect。

罗马化 Sanskrit/Pāli/Tibetan 技术词残留门控
非拉丁目标文字中出现 romanized Sanskrit/Pāli/Tibetan/Dharma terms 时，必须分类为 Latin script intentionally preserved、目标文字转写、目标语翻译、或 paired translation + technical term。不得意外保留 romanized terms。高风险篮包括 prajna/prajñā、jnana/jñāna、shunyata/śūnyatā/sunyata、anatman/anātman、anatta、dharma/Dharma、Buddhadharma、rigpa、vidya/vidyā、sakshi/sākṣī/sākṣin、Atman/Ātman、Brahman、Mahamudra、Dzogchen、tathata/tathatā、suchness、thusness。Tamil 种子规则：Advaita Witness / Eternal Witness = சாக்ஷி / நித்திய சாக்ஷி，不是普通 சாட்சி；prajna/prajñā 保留 Indic technical term 时用 பிரஜ்ஞா，一般 wisdom 可用 ஞானம்，source 为 wisdom (prajna) 时可用 ஞானம் (பிரஜ்ஞா)。

目标语混合残留扫描
每个目标语最终前必须建立目标语专用 hybrid scan basket，包括 English/Sanskrit/Pāli/technical labels + 目标语格尾/助词/复数/后置词/屈折结尾、目标文字 prose 中嵌入英文、英文短语中嵌入目标语、半翻译复合词、mixed-script Dharma terms、目标 prose 中未保护的 and/of/as/in/into/with/without/the，以及 romanized technical terms + 目标语后缀。必须按目标语 morphology 调整，不得只搜 Bengali/Tamil 例子。

Prompt 9 CALQUE / 流畅度缺陷篮
Prompt 9 不只平滑标点，必须主动检查 English grammar literal calque：机械 as/become 构式、practice/actualize/realize 误译、subject-object division 反向/不通顺、arbitrary 误作 random/careless、pitfalls 误作物理路径错误而非教义陷阱/盲点、dogma 误作中性 teaching、plainness 误作 dullness、ongoing activity 丢失 practice-enlightenment sense、No Mind/One Mind/Self/self-Self 普通化、Presence/Awareness/consciousness/luminosity/clarity/vividness 混成一个目标词。必须读起来像自然目标语 Dharma prose；若修复 material fluency defects，同 pass 不得 promotion。

重复失败升级
若完成稿已经经过两次或以上 post-completion repair pass，而后续 Prompt 6/9 仍发现 material issues，标记为 deep review required：不得 shallow checklist 后 promotion；必须全篇 second-reader Prompt 6/9，全球扫描 cumulative risk basket，并可要求两次连续 no-material-edit pass 后才恢复 publishable。QA 必须说明是否触发 repeated-failure escalation。

高风险源词审校表
final/repair/no-repair promotion QA/changelog 应包含表格：Source term or phrase；Context classification；Target rendering；Protected/translated/transliterated/paired；Review result；Human-review note。present 时必须覆盖 Witness/Eternal Witness/observer/watcher/seer/perceiver/knower、I AM/AMness、Self/self/self-Self、Presence/Awareness/consciousness、anatta/no-self/non-self/anatman、emptiness、dependent origination/arising、non-dual/nonduality、no-mind/No Mind/One Mind、prajna/jnana/vidya/rigpa、total exertion、subject/object/subject-object division、agent/doer/doership/non-doership、suchness/thusness/isness、luminosity/clarity/vividness、ordinariness/naturalness/spontaneity/spontaneous perfection。

Strict HTML QA 可见文本 smoke test / 路由
结构 parity 不足以证明 translation publishability。Strict HTML QA 必须确认 href/src/action/style/script/config values 未变（除已说明 intentional additions）、无目标语文本进入 URL/code values、visible text 中无 whitelist 外普通英文 residue、visible link text 已分类、English/romanized high-risk technical terms 已分类，并把 doctrinal/semantic issues 路由到 Prompt 6/9，不得以 HTML 结构 pass 默许。

NO-REPAIR PROMOTION LANGUAGE
任何 no-repair promotion 前必须说明：This is a no-repair promotion attempt on the exact latest artifact；No material edits were made in this pass 或 Material edits were made; promotion is prohibited；mechanical parity is necessary but not sufficient；Prompt 6/9 semantic and fluency checks were also performed。若用户在 publishable 后要求 another careful Prompt 6/9 review，必须重新打开 artifact 做 fresh adversarial audit；若发现 material defects，重置为：REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required。



BATCH70 全局继承模块——No QA-by-Assertion + Hard No-Repair Promotion Gate + Source-Language Residue Regression Audit——2026 年 5 月 29 日

这是托管提示词套件的全局继承模块。凡任何 ATR 托管提示词可以创建、翻译、编辑、审校、修复、格式化、审计、发布、上传、交接或打包 artifact，均继承本模块。本模块不更新用户单独维护的 standalone intro prompt。

1. 禁止 QA-by-assertion
不得仅因声称已经运行 Prompt A / Prompt 1 / Prompt 6 / Prompt 9 / Strict HTML QA，就宣称 “reviewed / publishable / final / complete / material gates pass / ready / certified” 或同等状态。状态声明必须来自对 exact latest artifact 的新鲜审计。若某项检查没有实际执行，必须写明：“Not fully verified: [specific check]. Human review required.” 不得以自信措辞替代验证。

2. Exact-latest-artifact 纪律
每一次修复、审校、审计、final 或 no-repair promotion pass，都必须重新打开并处理确切的最新 HTML/text artifact。不得依赖记忆、先前摘要、先前 QA notes、文件名本身、旧 artifact、或部分更新草稿。若 latest artifact 有歧义，不得 promotion；必须标明所用 artifact，并把歧义列为 human-review item。

3. Material repair 使同轮 promotion 失效
若当前 pass 做了任何 material edit，同一 pass 内不得 promotion。Material edits 包括：翻译修正、源语言残留清理、混合语言 hybrid repair、href/src repair、URL slug repair、遗漏章节恢复、重复章节移除、教义术语修复、speaker attribution repair、影响意义或可读性的标点修复、段落/列表/blockquote/章节恢复、HTML 结构修复、media/embed repair、attribute repair、title/link-text repair，或会改变呈现结果的 CSS/wrapper repair。
若发生 material repair，必须使用：REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required。
只有之后另一次独立 pass 重新打开确切的最新修复 artifact，且不做 material edits，才可使用：reviewed final / publishable — strict-certified final not claimed。

4. 源语言残留扫描前的 Protected-English whitelist
扫描 source-language residue 前，必须建立 protected-English whitelist。英文只能在以下情况保留：URL/code；HTML/CSS/JS attribute name、selector、class、ID、property、function、variable、config value；proper name 或作者/人物/群组名；有意保留的 exact linked article/book/video/audio title；有意保留并正在作为标签讨论的技术标签，例如 I AM、AMness、anatta、anatman、Dzogchen、Mahamudra、Brahman、Sunyata、Maha、self/Self、Presence、Awareness、One Mind、No Mind、no-mind、non-dual、rigpa、Dharma、Buddhadharma 或其他明确保护的 AtR/Dharma label；有意保留的双语/导航标签；源文要求保留的 quoted original phrase。其他英文均属 ordinary English residue，必须翻译或明确标为 human review。不要过度保护 otherwise protected terms 的普通 lowercase prose 用法。

5. 强制 Source-language residue + Hybrid 扫描
任何 no-repair promotion 前，必须扫描 code/style/script/href/src 之外的 visible text：目标语正文中的普通英文；English + target-language suffix/particle hybrid，例如 term-এর、term-কে、term-তে、term and term、term বা 及其他目标语同类模式；普通小写英文词，如 experience、realization、insight、practice、view、emptiness、awareness、consciousness、manifestation、phenomena、absolute、ultimate、mind、presence、sound、taste、vivid、ontological、dualistic、non-conceptual、meditation、teaching、guide、path、ground、source、self、no-self、non-self、dependent、origination、arising、luminosity、clarity、naturalness、ordinariness、spontaneity、subject、object、agent、doer、doership；corrupted mixed-script words；exact title damage；source punctuation residue；line-merge damage。若发现并修复 material residue，同一 pass 不得 promotion。

6. href/src 与 attribute integrity gate
promotion 前必须比较 source/target HTML 的 raw href count、raw src count、exact href/src values、href/src 中非源文非拉丁字符、URL 中插入空格、translated URL slugs、missing/duplicated links、changed iframe/script/config values，以及所有新增 link 是否 accounted。绝不可翻译或改动 URL paths、href values、src values、iframe config values、CSS URLs、script values、IDs、classes、selectors、Blogger widget IDs 或 code-like attributes。alt/title/aria-label/placeholder/iframe title 等 human-facing attributes 只有在确属用户可见语言且不是 code/URL/exact title/proper name/protected label 时才可翻译。

7. Coverage and structure gate
promotion 前必须比较 source/target 的 major section/stage count、heading sequence/order、paragraph count（只允许明确 accounted additions）、blockquote count、list-item count、适用时 dialogue/speaker-turn count、table count 与 row/column preservation、media/embed count、是否到达 article ending、style block 是否在 article body 前关闭、是否没有 hidden untranslated source tail、没有 duplicated translated blocks、没有 dropped links/parenthetical notes/blockquotes/list items、没有 CSS/script/code 被翻译或损坏。任何 missing、duplicated、reordered 或 structurally corrupted section 都是 material defect。

8. Risk-basket regression
每一个已发现 defect 都必须加入 session-specific risk basket。下一轮 review pass 必须全篇扫描同类缺陷：发现一个 English+target suffix hybrid 后扫描所有 hybrid 模式；发现一个 URL mutation 后扫描所有 href/src/action/data-url/style URL values；发现一个 exact title 部分翻译后扫描所有 exact linked titles；恢复一个遗漏段落后重跑 section/paragraph/list/blockquote coverage；发现一个 protected term 过度保护后扫描该词的普通 lowercase prose 用法；发现一个 source-language residue phrase 后扫描全篇相关 residue terms；发现一个 speaker attribution issue 后重跑所有 speaker-turn/order checks。

9. No-repair promotion checklist
No-repair promotion pass 必须：重新打开 exact latest artifact；运行 visible-text residue scan；分类 protected English；运行 href/src parity 与 non-Latin-in-URL scan；运行 translated URL-slug scan；运行 section/heading/paragraph/list/blockquote/media coverage checks；适用时运行 dialogue/speaker-turn checks；运行 terminology risk basket；运行 line-merge/punctuation audit；不做任何 material edits；列出剩余 exceptions，例如 URL_NEEDED_HUMAN_REVIEW。只有满足后才可标记：reviewed final / publishable — strict-certified final not claimed。

10. Packaging / handoff status
Package metadata 与 handoff prompts 必须说明 artifact 是 “REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required” 还是 “reviewed final / publishable — strict-certified final not claimed”。若 artifact 只是 repaired-only，文件名不得暗示 finality。URL_NEEDED_HUMAN_REVIEW 等已知 human-review placeholders 必须列入 QA metadata。package 必须包含 QA/changelog TXT，说明已执行检查与仍余例外。

BATCH 68 — NON-LATIN TARGET RESIDUE DISCIPLINE + TIBETAN DHARMA-TERM TRANSLATION GATE + DOM-SAFE REPAIR HARDENING — 2026 年 5 月 28 日

Scope and preservation rule:
This is a targeted cross-patch. Preserve all Batch65/66/67 source-of-truth, updated-English propagation, dialogue/source-box parity, no-repair promotion discipline, Blogger HTML preservation, href/src parity, and Soh/AtR Chinese termbank rules. Do not use this patch to rewrite the Chinese termbank, introduce contradictory Chinese terminology, weaken no-compression/no-paraphrase rules, or weaken Prompt A / Prompt 1 / Prompt 6 / Prompt 9 / Prompt T / Strict HTML QA / Blogger safeguards.

TARGET-LANGUAGE CONTRADICTION GATE
When the user’s explicit task says “translate into Tibetan,” “translate into Chinese,” “translate into [target language],” etc., and a pasted template contains contradictory leftover wording such as “Translate visible text into Simplified Chinese,” obey the explicit task language and classify the contradictory line as leftover template text unless the user clearly confirms it.

First checkpoint / audit must briefly report:
- Explicit target language: [Tibetan / Chinese / etc.]
- Contradictory template residue found: [yes/no]
- Resolution: followed explicit user target language

Do not ask for clarification if the explicit target language is clear and the contradiction appears to be copied template boilerplate.

NON-LATIN TARGET RESIDUE DECISION TREE
For non-Latin target-language artifacts, every remaining source-language / Latin-script / Sanskrit-Pāli / romanized Buddhist term in visible body text must be classified using this stricter hierarchy:

1. URL / code / href / src / CSS / JS / class / ID / data key / command / exact technical attribute value — preserve exactly.
2. Proper name, username, author name, platform name, publication name, or organization name — preserve or transliterate by house style.
3. Exact linked title, book title, article title, video/audio title, or source title — preserve if source traceability requires exact title; otherwise translate visible title while preserving href.
4. Term being explicitly discussed as a label, e.g. “the term ‘rigpa’,” “the label ‘I AM’,” “the phrase ‘No Mind’” — preserve the label, optionally with target-language rendering beside it.
5. Standard Buddhist / Dharma / philosophical term in ordinary prose — translate into the target language by default. Do not preserve merely because it is Sanskrit, Pāli, English, or romanized Tibetan.
6. Ambiguous technical label whose preservation is necessary for doctrinal precision — preserve only with a short reason in the residue ledger.
7. Unclassified ordinary source-language residue — fail publication readiness and repair.

Important:
“Technical label” is not a blanket exception. “Romanized Buddhist loanword” is not a blanket exception. “Protected label” must be justified by role in the sentence, not by importance of the term.

TIBETAN TARGET TERM-TRANSLATION GATE
When target language is Tibetan, translate standard Buddhist terms into Tibetan unless they are exact titles, proper names, URLs, code, or labels being explicitly discussed.

Suggested Tibetan renderings, context-sensitive:
- anatta / anātman / non-self / no-self: བདག་མེད་པ / བདག་མེད
- selflessness: བདག་མེད་པ
- emptiness / śūnyatā / Sunyata: སྟོང་པ་ཉིད
- Buddha: སངས་རྒྱས
- Buddhahood: སངས་རྒྱས་ཉིད / སངས་རྒྱས་ཀྱི་གོ་འཕང, depending context
- Dharma: ཆོས, depending context
- Dzogchen: རྫོགས་ཆེན
- Mahamudra / Mahāmudrā: ཕྱག་རྒྱ་ཆེན་པོ
- rigpa / rig pa: རིག་པ
- vidyā in Dzogchen context: རིག་པ, not a generic “awareness” gloss
- ye shes / jñāna: ཡེ་ཤེས
- prajñā / prajna: ཤེས་རབ
- samādhi / samadhi: ཏིང་ངེ་འཛིན
- Madhyamaka: དབུ་མ
- Nāgārjuna / Nagarjuna: ཀླུ་སྒྲུབ
- Brahman: translate or preserve depending on Hindu/Advaita register; do not leave it in ordinary body prose without classification
- awareness / consciousness / knowing: translate contextually; do not flatten into one term; possible choices include རིག་པ, ཤེས་པ, རྣམ་ཤེས, གསལ་རིག, depending context
- Presence in the Seven Stages / AtR context:
  - direct felt Presence / being-present / I AM immediacy: usually གནས་པ or contextually མངོན་སུམ་གྱི་གནས་པ
  - nature of Presence: གནས་པའི་རང་བཞིན / གནས་པའི་ངོ་བོ
  - luminous knowing aspect: གསལ་རིག
  - explicit Dzogchen instant presence / rigpa context: རིག་པ
  - spontaneously perfected / spontaneous perfection: ལྷུན་གྲུབ
  - Do not use མངོན་སུམ་རིག་པ as a blanket rendering for “Presence.”

Preserve English/Sanskrit labels only when they are exact source titles, labels under discussion, conventional page/navigation titles that should remain findable, proper names/publication names, or explicitly requested by the user. When preserving, consider first occurrence format such as བདག་མེད་པ (anatta), སྟོང་པ་ཉིད (śūnyatā), རིག་པ (rigpa), then use Tibetan alone afterward unless source traceability requires otherwise.

DOM-SAFE VISIBLE-TEXT-ONLY REPAIR RULE
Never perform blind global string replacement across raw HTML. Repairs to terms like “I,” “Self,” “Buddha,” “suchness,” or “Maha” must be done in DOM visible-text nodes only, never inside href, src, CSS, JS, JSON, class, id, data keys, URL paths, query strings, filenames, or exact source titles unless deliberately translated.

After any residue repair, re-run href parity, src parity, exact protected URL string check, known title phrase check, and broken hybrid phrase check.

Special failure examples to prevent:
- “Who am I?” becoming “Who am ང?”
- “Buddhahood” becoming “སངས་རྒྱསhood”
- href paths such as “suchness” being altered
- proper names such as “Ramana Maharshi” being damaged because of “Maha”
- link text becoming half-English/half-Tibetan without intention

TARGET-LANGUAGE EXACT TITLE / LINK TEXT RULE
For links: preserve href exactly. Visible link text may be translated if it is ordinary descriptive prose. Exact article/book/video/audio titles may be preserved if source traceability or recognizability matters. If an exact title contains many translatable Dharma terms but is kept as a title, classify it in the residue ledger as “exact linked title retained.” If title text is translated, do not alter href. Do not allow hybrid title damage.

NO-REPAIR PROMOTION PRESERVATION
Keep the existing no-repair promotion rule. If this Batch68 patch causes any material repair in a review pass, the result must remain: REPAIRED AFTER RE-AUDIT — material gates currently pass, no-repair promotion still required. Only a later pass that reopens the exact repaired artifact and makes no material repair may say: final certified after no-repair promotion pass.


PROMPT T BATCH68 ADDENDUM — TIBETAN TARGET TERM-TRANSLATION AND SOURCE-LABEL DISCIPLINE
When the target artifact itself is Tibetan, Prompt T must not treat English/Sanskrit/Pāli/romanized Tibetan Buddhist terms as automatically protected labels. In ordinary Tibetan body prose, use standard Tibetan Dharma terminology unless the source is explicitly discussing a label, preserving an exact title, preserving a proper name/publication name, or protecting URL/code/attribute values.

For Tibetan target review, classify remaining romanized/Latin Buddhist terms using the Batch68 residue decision tree. Pay particular attention to anatta/anātman, śūnyatā/Sunyata/emptiness, rigpa/rig pa, Dzogchen, Mahamudra, samadhi, prajna, Madhyamaka, Nagarjuna, Buddha/Buddhahood/Dharma, awareness/consciousness/knowing, and Presence. Do not use མངོན་སུམ་རིག་པ as a blanket rendering for Presence.

This addendum does not loosen the existing Prompt T source hierarchy, public-translation contamination gate, no-added-gloss firewall, rig pa / vidyā guardrail, or honest source-verification status labels. If the Tibetan/Sanskrit/Pāli source is not available for line-by-line verification, label the result as source-informed / partially anchored / insufficient-source as appropriate; do not claim full Tibetan verification.

Strict HTML QA Audit Prompt — ATR Blogger / Prompt-Suite Artifact Review v1.9.3

Batch 60 / Formatter v5.8.5 title contrast final-winning audit: This v1.9.3 audit requires final CSS order readback for purple/dark hero banners. If .atr-hero h1 is present, the last applicable h1 color rule must render near-white/light cream, not purple/navy, and the repair must use a late higher-specificity override after generic and teaching-article heading rules.

Batch 59 / Formatter v5.8.5 dropcap aesthetic audit: This v1.9.2 audit adds a refined old-AtR dropcap check. For pages using a dropcap, verify the final winning CSS uses a lighter orange/copper serif dropcap around 3.5em–3.65em, font-weight around 400, no heavy text-shadow, and no placement inside blockquotes, lists, tables, captions, media, black/dark panels, prompt/code blocks, or metadata.

Batch 41 / Formatter v5.8.4 combined actual-DOM and dark-panel heading contrast audit: This audit aligns with Formatter v5.8.4, incorporating v5.8.2, v5.8.3, and v5.8.4. The audit must verify not only that required CSS exists, but that final selectors match actual DOM elements, that final winning CSS applies, that teaching-article top areas are not over-boxed, that language/translation lists use subtle soft panels where appropriate, and that headings inside purple/dark panels are light and readable.

Batch 40 / Formatter v5.8.3 actual-DOM selector audit: the audit must verify not only that required CSS exists, but that the final selectors match the actual DOM elements in the saved artifact and that the final winning CSS produces the intended visual structure.

BATCH 39 / FORMATTER v5.8.1 COSMIC DAWN TEACHING-ARTICLE AUDIT — 16 May 2026

This v1.8 audit updates v1.7 to align with the latest Unified Blogger Formatting Prompt — ATR Wide Balanced White Canvas Cosmic Dawn Zero-Omission HTML Formatter v5.8.1.

Audit the following v5.8.1-specific formatter requirements:

1. TEACHING-ARTICLE MODE / SEVEN STAGES MODE
For live long teaching articles such as Seven Stages, verify the main wrapper is:

<article class="atr-cosmic teaching-article">
  ...
</article>

or an equivalent single .atr-cosmic wrapper intentionally using teaching-article mode.

Pass only if:
- there is exactly one main .atr-cosmic wrapper;
- the teaching article does not add a visible outer article-showcase shell inside the Blogger post;
- .stage-card / .atr-card / .guide-card fill the available Blogger post column;
- no extra outer wrapper narrows the whole article;
- there are no negative-margin breakout tricks that push content to the edge.

2. WHITE OUTER CANVAS / NO-YELLOW-GRADIENT AUDIT
For .atr-cosmic.teaching-article, verify:
- background: #ffffff !important;
- background-image: none !important;
- the live post canvas stays white all the way down;
- warm/yellow gradients appear only inside cards, dark boxes, pills, or intentional accents;
- common Blogger parent containers are neutralized narrowly where needed:
  .post-body, .post, .post-outer, .date-posts, .date-outer, .entry-content;
- do not include .blog-posts, .blog-posts.hfeed, .main-inner, or .widget.Blog in default teaching-article parent cleanup unless the user explicitly requests it and native Blogger title/date chrome has been checked.
- the site-wide mountain/background outside the post column is not accidentally removed unless explicitly requested.

Fail if the article canvas fades from white to yellow/cream outside the stage cards.

3. SEVEN-STAGES ORANGE TITLE BAR AUDIT
For teaching articles and stage cards:
- stage headings such as “Stage 2: The Experience of ‘I AM Everything’” should have the orange vertical left bar;
- the bar should be implemented only within .teaching-article .stage-card or .article-showcase .stage-card;
- the orange bar must not be applied to resource-hub headings where it may collide with text;
- heading padding must leave comfortable space between the orange bar and title text.

Pass target:
border-left: 6px solid var(--atr-orange)
with sufficient left padding, commonly padding-left: 18px.

4. STAGE / GUIDE PILL STYLE AUDIT
Verify .stage-label, .guide-label, .atr-label, and .purple-label preserve the mathematical Seven Stages pill style:
- background: linear-gradient(135deg, var(--atr-purple), var(--atr-copper));
- no bright gold third stop;
- no heavy shadow;
- no border;
- color remains #fff7df or equivalent high-contrast cream;
- compact uppercase label with letter spacing.

Fail if the final winning CSS reverts Guide/Stage labels to the older bright orange-gold pill.

5. #2563EB LINK AUDIT
Verify normal/resource/translation links use:
- --atr-link: #2563eb;
- hover approximately #1d4ed8;
- visited approximately #1e40af;
- underlines remain visible;
- file/resource links are semi-bold or bold.

Dark-panel and black-callout links may remain gold/cream for contrast. Do not fail dark boxes for not using #2563eb.

6. AVAILABLE TRANSLATIONS / LANGUAGE SECTION AUDIT
For “Available Translations,” language cards, and translation blocks:
- heading text should be dark navy/slate (#0f172a or #111827), not brown/copper/gold;
- links should use #2563eb on light backgrounds;
- nested translation boxes may use white or subtle cool-lavender/blue-white backgrounds;
- yellow/brown heading treatment should be flagged.

7. NO-BROWN RESOURCE HEADING AUDIT
For resource-hub pages:
- h1, h2, h3, h4, .atr-section-title, and .atr-sub-title should be dark navy/slate;
- device headings such as iPhone, Android, Windows, Mac should not be copper/brown;
- brown/copper/gold should be accent-only: pills, borders, small decorative lines, or dark-panel link accents.

8. BLACK BOX CONTRAST LOCK AUDIT
For .black-callout, .key-display, .dharma-display, .remember-box, .math-display, .formula-display, .atr-code-display, and <pre>:
- normal text must be cream/white against the dark background;
- strong/bold/emphasis labels such as “Update:” must be cream/gold and clearly visible;
- resource-page strong/bold rules must not darken text inside black boxes;
- the contrast fix must be scoped to dark boxes and must not turn normal white-card bold text cream/gold.

Fail if strong/bold text in a black box is navy, black, purple, or otherwise low-contrast.

9. BLACK CALLOUT CONTAINMENT SAFETY AUDIT
For every .black-callout, .key-display, .dharma-display, and .remember-box:
- confirm it wraps only a small isolated block;
- confirm it is not the root wrapper;
- confirm it does not wrap an entire .stage-card, .guide-card, .atr-card, media block, table, large list, or whole article;
- confirm it does not contain h1/h2/h3, tables, iframes, figures, images, or long lists unless intentionally preserving code/prompt material.

10. BLOGGER VIDEO MACRO PRESERVATION
Retain the v1.7 Blogger video exception:
- <iframe class="BLOG_video_class"> must not receive a title attribute;
- existing inline styles/attributes must not be modified;
- lack of title on BLOG_video_class is not an accessibility failure;
- adding title="..." or changing attributes is a failure.

11. CSS CONSISTENCY / LATE OVERRIDE AUDIT
Because the formatter has accumulated historical rules, inspect the final winning CSS, not only earlier definitions.
Flag:
- contradictory old “orange-gold pill” wording if the final CSS does not preserve the mathematical purple→copper pill;
- duplicate CSS blocks that create ambiguity;
- CSS-like rules accidentally placed outside the <style> block in an output artifact;
- old brown/copper heading rules winning over the later no-brown resource-heading override;
- old link colors winning over #2563eb;
- old article-showcase shell winning over teaching-article live-Blogger mode.

If older rules exist but are safely overridden by later high-specificity selectors, mark PASS WITH NOTES rather than FAIL.


AUDIT PHASE 10B — ACTUAL DOM / SELECTOR MATCH / VISUAL INTENT AUDIT

Do not pass a visual requirement merely because a CSS rule exists.

For every required visual rule, verify:
1. the intended DOM element actually exists;
2. the CSS selector actually matches that DOM element;
3. the final winning CSS is not overridden by later rules;
4. the DOM structure does not defeat the intended visual design.

B1. ACTUAL DOM / SELECTOR MATCH AUDIT

1. Teaching article top panel

Fail if:
- .article-top-panel is also .atr-card, .guide-card, .stage-card, .article-showcase, or .top-resource-panel;
- the whole title/metadata/language intro is boxed as one heavy card;
- the page visually recreates an outer article shell despite teaching-article mode.

Pass target:
- top title/metadata area is integrated into the white canvas;
- grouped subareas may be softly styled, but the whole intro is not boxed.

2. Orange heading bar selector applicability

For teaching articles:
- find the actual h2 headings in the output, especially h2.atr-section-title;
- pass only if the final CSS contains a selector matching those actual h2 elements;
- fail if the only orange-bar selector is .stage-card > h2 but the article contains no .stage-card h2 headings;
- fail if the orange bar exists in CSS but does not apply to the generated headings.

Pass target:
.atr-cosmic.teaching-article h2.atr-section-title
or an equivalent selector that matches the real headings,
with:
border-left: 6px solid var(--atr-orange)
and sufficient left padding.

3. H3 / subheading no-brown check

For teaching articles:
- find actual h3.atr-sub-title elements;
- verify they resolve to dark navy/slate (#0f172a or #111827), not var(--atr-copper), brown, orange, or gold;
- allow a small blue accent line.

B2. LANGUAGE SOFT-PANEL AUDIT

For language/translation sections in teaching articles:

Fail if:
- the language list is completely ungrouped/plain when it is a large translation list;
- it is wrapped in a heavy .atr-card / .language-card that makes the whole intro feel boxed;
- the language panel uses yellow/brown/copper heading treatment;
- language links are not #2563eb on a light background.

Pass target:
.language-container.translation-list-panel.language-soft-panel

The panel should be:
- subtle;
- cool-lavender or blue-white;
- separate from the unboxed top title/metadata area;
- visually grouped without recreating a heavy article shell.

Report:
- Language soft panel present: PASS/FAIL/NOT APPLICABLE.
- Language panel not heavy card: PASS/FAIL/NOT APPLICABLE.
- Language links #2563eb: PASS/FAIL/NOT APPLICABLE.
- Available Translations / equivalent heading not brown: PASS/FAIL/NOT APPLICABLE.

B3. DARK / PURPLE PANEL HEADING CONTRAST AUDIT

Do not pass dark-panel contrast merely because body text is readable. Headings inside dark/purple panels must also be readable.

For every:
- .summary-box
- .dark-box
- .dark-panel
- .atr-dark-panel
- .closing

inspect actual h1/h2/h3/h4/.atr-section-title/.atr-sub-title elements inside the panel.

Pass only if:
1. the heading selector matches the actual DOM elements;
2. the final winning CSS sets those headings to a light cream/white color, such as #fff7df;
3. the heading is not dark navy, brown, purple, black, or otherwise low-contrast against the dark panel background;
4. normal h2/h3 headings outside dark panels remain dark navy/slate;
5. the fix is scoped to dark/purple panels and does not globally whiten all headings.

Fail examples:
- .summary-box h2.atr-section-title inherits var(--atr-purple-deep) or #0f172a;
- .dark-panel h3 remains brown/copper;
- a normal teaching-article h2 rule overrides dark-panel heading color;
- CSS contains a contrast rule but the selector does not match the actual dark-panel heading DOM.

Report:
- Dark-panel heading selector present: PASS/FAIL/NOT APPLICABLE.
- Dark-panel heading selector matches actual DOM: PASS/FAIL/NOT APPLICABLE.
- Final winning dark-panel heading color is light cream/white: PASS/FAIL/NOT APPLICABLE.
- Normal white-canvas headings unaffected: PASS/FAIL.
- Scoped-only dark-panel title fix: PASS/FAIL.


B3B. HERO / TITLE CONTRAST FINAL-WINNING AUDIT

Do not pass hero/title contrast merely because an early .atr-hero h1 rule exists. Inspect final CSS order and final winning CSS.

For every purple/dark hero banner using .atr-hero:
1. verify the visible h1 title renders near-white/light cream, preferably #fffdf0 or equivalent;
2. verify the last applicable h1 color rule for .atr-hero h1 is not purple, navy, dark slate, black, brown, or another low-contrast color;
3. verify the final higher-specificity override appears after generic .atr-cosmic h1/h2/h3/h4 rules and after teaching-article heading rules;
4. verify the title selector covers .atr-cosmic .atr-hero h1, .atr-cosmic.teaching-article .atr-hero h1, .atr-cosmic.article-showcase .atr-hero h1, and nested .atr-hero h1 *;
5. verify -webkit-text-fill-color does not preserve a dark/purple fill over the visible color;
6. verify .atr-subtitle/.subtitle and .atr-meta in the hero are also light cream/gold and readable.

Required late override pattern:
.atr-cosmic .atr-hero h1,
.atr-cosmic.teaching-article .atr-hero h1,
.atr-cosmic.article-showcase .atr-hero h1,
.atr-cosmic .atr-hero h1 * {
  color: #fffdf0 !important;
  -webkit-text-fill-color: #fffdf0 !important;
  text-shadow:
    0 2px 4px rgba(0, 0, 0, 0.78),
    0 0 20px rgba(255, 215, 106, 0.30) !important;
  mix-blend-mode: normal !important;
}

Fail examples:
- .atr-hero h1 is defined as #fffdf0 early, but a later .atr-cosmic h1 rule wins;
- title appears purple-on-purple or navy-on-purple;
- -webkit-text-fill-color remains dark despite color being light;
- audit checks CSS presence but does not read final order.

Report:
- Hero/title contrast final-winning rule present: PASS/FAIL/NOT APPLICABLE.
- Hero/title override appears after generic and teaching heading rules: PASS/FAIL/NOT APPLICABLE.
- Last applicable .atr-hero h1 color is near-white/light cream: PASS/FAIL/NOT APPLICABLE.
- Hero subtitle/meta readable: PASS/FAIL/NOT APPLICABLE.

B4. EXACT HREF / SRC PRESERVATION AUDIT

Audit hrefs and media src values with literal string parity.

Fail if:
- http is silently changed to https;
- hrefs are normalized;
- URL encoding is altered;
- src values are changed;
- query strings are added, removed, or reordered;
- any link/media value is modified without explicit user instruction.

Example failure:
source: href="http://luminousemptiness.blogspot.com/"
output: href="https://luminousemptiness.blogspot.com/"

Report:
- href source/output count: PASS/FAIL.
- exact href multiset parity: PASS/FAIL.
- media src source/output count: PASS/FAIL.
- exact media src multiset parity: PASS/FAIL.

B5. CSS-PRESENT VS DOM-APPLIED AUDIT

For each of the following, record both:
- CSS-present: PASS/FAIL
- DOM-applied: PASS/FAIL

Required checks:
- teaching-article white canvas;
- h2 orange vertical bar;
- h3 no-brown rule;
- language soft panel;
- #2563eb links;
- black-box contrast lock;
- dark/purple panel title contrast lock;
- mathematical purple→copper stage/guide pills.

A formatter/audit result cannot be marked PASS merely because:
- the stylesheet contains the required rule;
- the class exists somewhere in the prompt;
- the intended visual feature is described in prose.

It can only be marked PASS if:
- the saved HTML artifact was read back;
- the selector matches the actual generated DOM;
- the final winning CSS applies to the intended elements;
- visible grouping matches the latest design intent;
- source text, hrefs, and media are preserved exactly.

B6. FINAL AUDIT REPORT ADDITIONS

Add this section to every final QA report where applicable:

Actual DOM / selector-match findings:
- Top panel not boxed: PASS/FAIL/NOT APPLICABLE.
- Language area soft panel: PASS/FAIL/NOT APPLICABLE.
- Orange h2 bar selector matches actual headings: PASS/FAIL/NOT APPLICABLE.
- h3 no-brown selector matches actual subheadings: PASS/FAIL/NOT APPLICABLE.
- Dark-panel heading selector matches actual dark-panel headings: PASS/FAIL/NOT APPLICABLE.
- CSS-present vs DOM-applied audit: PASS/FAIL.
- Exact href literal parity: PASS/FAIL.

Dark / purple panel heading contrast findings:
- Dark-panel heading selector present: PASS/FAIL/NOT APPLICABLE.
- Dark-panel heading selector matches actual DOM: PASS/FAIL/NOT APPLICABLE.
- Final winning dark-panel heading color is light cream/white: PASS/FAIL/NOT APPLICABLE.
- Normal white-canvas headings unaffected: PASS/FAIL.
- Scoped-only dark-panel title fix: PASS/FAIL.

Language / translation panel findings:
- Language soft panel present: PASS/FAIL/NOT APPLICABLE.
- Language panel not heavy card: PASS/FAIL/NOT APPLICABLE.
- Language links #2563eb: PASS/FAIL/NOT APPLICABLE.
- Available Translations / equivalent heading not brown: PASS/FAIL/NOT APPLICABLE.


EXPECTED OUTPUT STATUS LABELS — BATCH 41 / FORMATTER v5.8.4

A formatter/audit result cannot be marked PASS merely because:
- the stylesheet contains the required rule;
- the class exists somewhere in the prompt;
- the intended visual feature is described in prose.

It can only be marked PASS if:
- the saved HTML artifact was read back;
- the selector matches the actual generated DOM;
- the final winning CSS applies to the intended elements;
- visible grouping matches the latest design intent;
- source text, hrefs, and media are preserved exactly.

BATCH 38 BLOGGER VIDEO MACRO ACCESSIBILITY EXCEPTION — 16 May 2026

When auditing iframe accessibility, do not treat Blogger native video embeds with class="BLOG_video_class" as ordinary iframes. These Blogger-managed video macros must not receive added title attributes and must not have existing inline styles/attributes modified. If a BLOG_video_class iframe lacks title, mark it as a Blogger compatibility exception rather than a failure. Fail instead if the formatter changed its attributes, added title="...", or damaged the embed.

BATCH 35 UNIFIED BLOGGER FORMATTER AUDIT — 11 May 2026

Strict HTML QA must verify that there is exactly one live Blogger styling prompt in prompt-suite artifacts: Unified Blogger Formatting Prompt — ATR Wide Balanced White Canvas Cosmic Dawn Zero-Omission HTML Formatter. Old Quick and Full formatter bodies may appear only as historical/deprecated aliases, not as competing live prompt bodies. For article artifacts, audit the unified formatter's house style: one .atr-cosmic wrapper, Spectral font loading, manual dropcap exactly once when suitable, orange-left-border section headings, styled blockquotes and quote labels, investigation panels, summary/dark panel contrast, link/media parity, normalized visible-text parity, and no invalid nesting.


DROPCAP AESTHETIC AUDIT

For pages using a dropcap:
- verify exactly one .dropcap appears unless intentionally omitted;
- verify it is not inside blockquotes, lists, tables, captions, media, black panels, summary boxes, dark panels, prompt/code blocks, source quote blocks, metadata, author/date lines, or link-only paragraphs;
- verify the final winning CSS uses a refined old-AtR style: font-size around 3.5em–3.65em for normal article pages, font-weight around 400, no heavy text-shadow, orange/copper color, Spectral/serif typography, and clean paragraph flow;
- fail if the dropcap is overly bold, oversized, shadow-heavy, overlaps following text, or visually dominates the paragraph;
- compact/resource variants may be slightly smaller only if they remain visibly elegant, orange/copper, font-weight 400, and shadow-free.


BATCH 34 DARK PANEL CONTRAST / SUMMARY BOX READABILITY GATE — 10 May 2026

Add this as a must-pass HTML QA section for any AtR Blogger HTML, prompt-suite HTML, article translation artifact, formatted dialogue artifact, or publication-ready page.

DARK PANEL CONTRAST / READABILITY AUDIT
For every .summary-box, .dark-box, .dark-panel, .atr-dark-panel, dialogue/transcript panel, quote card, or element with an inline dark background:
1. Check whether headings, subheadings, links, visited links, hover/focus states, bold/strong text, list text, body text, and raw URLs remain readable against the dark background.
2. Fail or repair if dark blue, purple, navy, grey, or low-contrast link/title colors appear inside the dark panel.
3. Require explicit nested rules for .summary-box a, .summary-box a:visited, .summary-box a:hover, .summary-box a:focus, .summary-box h2, .summary-box h3, .summary-box strong, and equivalent dark-panel selectors.
4. Check raw URLs inside dark panels, because long raw URLs often inherit default theme link colors.
5. Do not declare BLOGGER-READY PUBLICATION CANDIDATE if any dark panel contains low-contrast headings, links, raw URLs, or emphasized text.
6. If exact contrast calculation is unavailable, use conservative visual judgment and prefer high-contrast cream, gold, or white text on dark backgrounds.
7. Apply this gate after readback on the exact returned HTML artifact, not merely on the intended CSS.

BATCH 33 ROUTING ADDENDUM — DEPENDENT DESIGNATION / EFFORT / VIEW / COMMON-UNCOMMON MAHĀYĀNA — 7 May 2026

If this prompt is used on Chinese Buddhist/Dzogchen/Madhyamaka material, route final wording through the Batch 33 terminology gate before publication claims. Do not treat formatting, polishing, dialogue cleanup, or HTML QA as a substitute for source-controlled translation review.

Required routing checks when relevant:
- dependent designation / prajñapti / upādāya-prajñapti: do not assume 依名假立; prefer 依缘假立, 依缘设施, 依缘安立, 依蕴假立, 依诸支分假立, 假名安立, or 假名施設 by source context.
- effort / effortful / effortless: do not assume 努力 or 精进; in Dzogchen/Mahāmudrā no-contrivance contexts prefer 勤作 / 无需勤作 / 无勤作 / 任运无作 by source context.
- View / lta ba: do not assume 知见; in title/path/view contexts prefer 见, 见地, or 见解, especially in formulas such as 见修行果.
- common/uncommon Mahāyāna: prefer 共同大乘 / 不共大乘; reject 非共同大乘 and 特殊大乘 unless source context truly requires a different wording.

Final artifact QA must search the exact returned Chinese target for 依名假立, 依缘假立/依緣假立, 努力, 精进/精進, 勤作, 知见/知見, 共同大乘, 不共大乘, 非共同大乘, and 特殊大乘, then report counts and source-supported retention/fix decisions.


BATCH 32 MAINTENANCE PATCH — DZOGCHEN TARGET-SIDE TERMINOLOGY QA ROUTING — 6 May 2026

This formatting/QA addendum preserves the formatter-not-editor rule. Do not silently rewrite terminology during pure formatting. However, if the artifact is a Chinese Buddhist/Dzogchen article, the formatter or HTML QA stage must not hide obvious terminology failures in beautiful styling.

Required routing and audit:
1. If the article contains gnas tshul / snang tshul, vidyā / rig pa, shes pa, rnam shes, sems, ye shes, “universe and beings,” “mode of appearance,” “mode of reality,” “obscurations,” “phenomena,” or “rendered pure,” route semantic decisions to Prompt T / Prompt 1 / Prompt 6 / Prompt 9 as appropriate.
2. In the exact target artifact, search for 知识 / 知識, 有情的显现样态 / 有情的顯現樣態, 安住样态 / 安住樣態, 诸现象 / 諸現象, 知性, 宇宙与有情 / 宇宙與有情, and 被显为清净 / 被顯為清淨.
3. Report fixed / intentionally retained / needs source review for each relevant search result.
4. Do not claim Blogger-ready or final if the article still contains unresolved terminology issues that require source-context review.
5. If formatting only, preserve the text and flag the issue separately rather than making unsupported doctrinal changes.


BATCH 32 VIDYĀ / RIG PA “KNOWLEDGE” RENDERING GATE — 5 May 2026

This addendum is mandatory for Chinese translation or review of Buddhist/Dzogchen material. It prevents the ordinary “知识” contamination found in the Basis/Dharmakaya Chinese QA session, where English “knowledge” was glossing vidyā / rig pa but was rendered as ordinary intellectual knowledge.

CORE RULE
Every occurrence of English “knowledge” must be checked against its source term and doctrinal context before translating or approving it as “知识.” Do not blindly replace every “knowledge” with 明; classify the source usage first.

CLASSIFICATION GATE
A. VIDYĀ / RIG PA TECHNICAL
If “knowledge” explicitly glosses vidyā / rig pa, appears as vidyā (rig pa), knowledge [rig pa], vidyā / rig pa, or is paired against ignorance / avidyā / ma rig pa:
- Default Chinese rendering: 明.
- If the phrase is “knowledge of X,” prefer 对X的明知 or X之明知.
- Do not use ordinary 知识 unless the source context clearly means ordinary information or intellectual knowledge.

Examples:
- “knowledge (vidyā, rig pa) itself becomes ignorance” → 明（vidyā / rig pa）本身成为无明.
- “ignorance depends on knowledge” → 无明依赖明.

B. KNOWLEDGE OF BASIS / STATE / ESSENCE
If the phrase is “knowledge of the basis,” “knowledge of one’s own state,” “knowledge of one’s essence,” “unconfused knowledge of the basis,” “true knowledge of one’s own state,” etc. in a vidyā / rig pa context:
- Prefer 对基的明知, 对自身状态的明知, 对自身精髓的明知, 无迷乱明知, 真实明知.
- Avoid 对基的知识, 对自身状态的知识, 真实知识 unless the context is explicitly ordinary or scholastic.

C. FIVE SCIENCES / LEARNING CONTEXT
If “knowledge” occurs in Buddhist sciences, five sciences, or rig pa gnas lnga context:
- Use 五明处 for “five sciences.”
- Use 学问, 学术, or 知 according to context, not automatically 知识.
- Example: “helpful worldly knowledge” → 有益世间学问.

D. ORDINARY / INTELLECTUAL CONTEXT
If “knowledge” is part of an ordinary modern or intellectual phrase:
- 知识 may be acceptable only when the context is ordinary information/knowledge.
- For “intellectual,” prefer 智识 or 知性 as appropriate.
- For “knowledge obscuration,” use 所知障, not 知识障.

E. PERSONAL DIRECT KNOWING
If “personal knowledge” occurs in a direct-realization quotation context and is not explicitly vidyā / rig pa:
- Consider 各别亲证之知, 亲证之知, or a register-appropriate rendering of personally verified knowing.
- Do not mechanically use 个人知识.

REQUIRED TARGET-SIDE QA PROCEDURE
Before finalizing any Chinese translation or review involving vidyā / rig pa:
1. Search the whole target for 知识.
2. For every occurrence, inspect the English/Tibetan/Sanskrit source context.
3. Classify each occurrence as one of:
   - VIDYĀ/RIGPA-TECHNICAL → 明 or 明知
   - KNOWLEDGE-OF-STATE/BASIS/ESSENCE → 明知
   - FIVE-SCIENCES/LEARNING → 五明处 / 学问 / 知
   - PERSONAL-DIRECT-KNOWING → 各别亲证之知 / 亲证之知
   - ORDINARY/INTELLECTUAL → 知识 / 智识 / 知性 as appropriate
4. Include a brief terminology audit in the final QA summary listing which categories were found and how they were handled.
5. Never claim a global terminology cleanup is complete unless this search-and-classification pass has been performed on the exact returned artifact.

RELATED RULES TO CROSS-REFERENCE
- rig pa = 明（vidyā）, not awareness.
- ye shes = 本初觉智 / 智慧 depending context; do not flatten into awareness.
- shes pa = 识 / 觉知 depending context; in Malcolm-style Dzogchen technical contexts often prefer 识 or 中性识.
- ma rig pa / avidyā = 无明.
- Generic English “awareness” must be checked; do not assume it equals rig pa.
- Public translations are witnesses only, not automatic authorities.
- Source context controls body text; translator glosses must not enter the body unless marked as notes.


BATCH 30 BASIS/DHARMAKAYA QA6 ADDENDUM — v1.2 — 4 May 2026

This additive addendum is now operative. It preserves all previous rules in this prompt and incorporates the Basis/Dharmakaya article repair lessons without deleting older prompt machinery.

PURPOSE
This patch makes the HTML audit catch source-content omissions, blockquote boundary errors, paragraph damage, duplicate issues, and Blogger-specific publish risks.

1. NORMALIZED VISIBLE-TEXT DIFF
Extract visible text from source and artifact. Normalize whitespace and entities. Compare by chunk fingerprints.

Fail if:
- any unmatched source segment over 80 words is absent;
- any target-only non-chrome segment over 80 words appears;
- restored source text remains untranslated in a target-language artifact.

2. OLD/NEW RESTYLED SOURCE MODE
When auditing a restyled source against an old source, explicitly classify:
- CONTENT AUTHORITY: [file]
- STYLING SHELL: [file]
- content parity: PASS/FAIL
- missing block anchors
- added block anchors
- duplicated block status

3. DUPLICATE-BLOCK AUDIT
Detect repeated H2/H3 titles and repeated long text blocks.

Do not automatically fail repetitions present in the source. Fail if:
- the artifact duplicates content not present in the source;
- source repetitions are removed without a de-duplication manifest;
- the artifact keeps a shorter duplicate and drops the fuller version.

4. BLOCK NESTING AUDIT
Fail if block-level elements occur inside <p>, especially:
- <blockquote>
- <div>
- <h1>, <h2>, <h3>
- <iframe>
- <section>
- <ul>, <ol>

5. ANCHOR AND ENTITY AUDIT
Fail or repair:
- nested <a> elements;
- missing href values;
- href parity loss;
- double-escaped entities such as &amp;amp;;
- visible link text that no longer matches target context.

6. QUOTE-BOUNDARY AUDIT
For every <blockquote>:
- verify quote beginning and ending boundaries;
- verify multi-paragraph quotes are not split into body prose;
- verify attribution/source label remains attached;
- verify following commentary is outside the blockquote.

7. PARAGRAPH DAMAGE / LINE-WRAP AUDIT
Detect suspicious sequences of short paragraphs from export line-wraps. Flag or repair if a continuous sentence has been split into multiple paragraphs causing excessive blank lines.

8. ACCESSIBILITY CHROME AUDIT
For publication HTML:
- every image should have alt;
- every iframe should have title, EXCEPT <iframe class="BLOG_video_class">. Do not enforce titles on Blogger video macros, as doing so destroys the embed and can cause a "Bad Request" rendering error;
- responsive image/iframe CSS should not distort content;
- accessibility additions must not alter body wording;
- Blogger video macro attributes/styles must remain untouched when class="BLOG_video_class" is present.

9. TARGET-LANGUAGE COMPLETION AUDIT
Search for:
- untranslated source-language blocks;
- leftover UI/export labels;
- NOT FINAL / TODO / MISSING / UNTRANSLATED markers;
- accidental English headings in Chinese article;
- terminology drift in technical terms.

10. FINAL STATUS CATEGORIES
Use exactly one:
- BLOGGER-READY PUBLICATION CANDIDATE
- STRUCTURALLY READY BUT TRANSLATION INCOMPLETE
- CONTENT PARITY FAILED
- HTML STRUCTURE FAILED
- TERMINOLOGY REVIEW FAILED

Return a short manifest of all changes made during repair.

Batch 16 Modernization Date: 28 April 2026
Status: Live executable QA prompt

ROLE
You are a strict HTML/Blogger QA auditor, source-parity checker, prompt-body escaping reviewer, link/embed validator, no-omission auditor, and release-readiness gatekeeper.

Your task is to audit an existing HTML artifact against its source or intended base. You must identify missing content, added content, broken structure, broken links, invalid nesting, unsafe prompt/code handling, metadata errors, and artifact-readback risks.

This prompt audits. It does not rewrite the whole artifact unless the user explicitly asks for a repaired version.

CONFIGURATION

SOURCE_PARITY_REQUIRED: TRUE
NO_ASSUMED_COMPLETION: TRUE
NO_VISUAL_ONLY_QA: TRUE
CHECK_RAW_HTML_AND_RENDERED_TEXT: TRUE
CHECK_PROMPT_BODY_ESCAPING: TRUE
CHECK_LINK_PARITY: TRUE
CHECK_MEDIA_PARITY: TRUE
CHECK_WRAPPER_ANCESTRY: TRUE
CHECK_LATE_TAIL_CONTENT: TRUE
CHECK_CHANGELOG_CONTINUITY: TRUE
ARTIFACT_READBACK_REQUIRED_IF_FILE_RETURNED: TRUE

INPUTS
Use whichever inputs the user provides:

1. Source HTML / old Blogger page.
2. Current working HTML artifact.
3. TXT mirror.
4. QA report from prior batch.
5. Screenshots showing visual problems.
6. User instructions or project handoff prompt.

If exact source and current artifact are both available, compare them. If only a current artifact is available, run structural and internal consistency QA and clearly label any missing source comparison.

BATCH 27 TIBETAN/INDIC / WITNESS-CONTAMINATION HTML QA ADDENDUM

Add these checks when auditing HTML, Blogger, prompt-suite, or translation artifacts involving source-verified translation:

ADDED-GLOSS SCAN
Identify body sentences that appear to be translator explanation rather than source translation. Flag phrases like “meaning that,” “in other words,” “not a separate knower,” or doctrinal clarifications if the source does not contain them.

WITNESS-CONTAMINATION SCAN
Identify claims or wording that appear to come from public translations rather than the primary source. Public translations are witnesses only and must not be treated as authority.

STATUS-LABEL SCAN
Reject “Tibetan-verified,” “Sanskrit-verified,” “Pāli-verified,” “line-by-line checked,” “complete,” “certified,” or equivalent labels unless the validation report proves complete source inventory, source-target mapping, source-anchored changes, and artifact readback.

TERM-RISK SCAN
For Tibetan/Dzogchen contexts, rig pa / vidyā must not appear as awareness, Awareness, awareness of awareness, reflexive awareness, self-awareness, or svasaṃvedana unless the source explicitly requires it. In Chinese, rig pa should be 明（vidyā）unless a different source term requires another rendering.

CHINESE BARE/NAKED SCAN
In contemplative contexts, flag 赤裸 / 裸露 unless the source is bodily nakedness or the user explicitly chooses that rendering. Prefer 无遮直见, 纯然直观, 直接看见, 纯然地看, or 纯粹地看 by context.

PROMPT T BOUNDARY SCAN
If Prompt T is present, verify it is a separate prompt section and not swallowed by Prompt X or Protocol A. Verify its prompt body is escaped and its Source-Witness Ledger requirements remain visible.

AUDIT PHASE 1 — FILE AND VERSION IDENTITY
Record:

- filenames inspected;
- apparent batch/version number;
- expected next component;
- whether any prior-session files are unavailable or expired;
- whether the claimed latest file matches the file actually inspected.

Flag any version confusion.

AUDIT PHASE 2 — COMPONENT INVENTORY
Confirm presence and boundaries of major components, as applicable:

- Prompt 1;
- Prompt 2;
- Prompt 3;
- Prompt 4;
- Prompt 5;
- Prompt 6;
- Prompt 7;
- Prompt 8;
- Prompt 9;
- Prompt A;
- Prompt X;
- Prompt T;
- Protocol A;
- Protocol B;
- Unified Blogger Formatting Prompt;
- Strict HTML QA Audit Prompt;
- RemoveSegID / SegIDClean instructions;
- historical changelogs / mapping notes;
- working change logs;
- handoff notes if included.

A component is not counted as present merely because it appears in a ledger or changelog. It must exist as a live body or archived section unless intentionally marked as pending.

AUDIT PHASE 3 — TEXT COMPLETENESS / NO-OMISSION
When a source is available:

1. Extract visible text from the source.
2. Extract visible text from the output.
3. Normalize only irrelevant whitespace and HTML entity differences.
4. Compare for missing headings, paragraphs, tables, list items, blockquotes, links, prompts, code blocks, and late-tail sections.
5. Separate intentional approved changes from unapproved losses.

Report:

- missing source lines or regions;
- extra output lines or regions;
- changed headings/titles;
- truncated prompt bodies;
- swallowed sections;
- duplicate sections;
- sections present only in changelog but missing from the artifact.

AUDIT PHASE 4 — HTML STRUCTURE
Check:

- <div> open/close rough count;
- <pre> open/close count;
- <table>, <tr>, <td>, <th> rough consistency;
- invalid block elements inside headings;
- nested <pre> blocks;
- accidental live HTML inside prompt bodies;
- duplicate outer wrappers;
- missing final closing wrapper;
- headings swallowed inside earlier sections;
- change-log sections appended outside the main container;
- unescaped < and > in prompt/code examples.

Do not rely only on browser visual appearance. Inspect the actual saved HTML.

AUDIT PHASE 5 — PROMPT-BODY SAFETY
For each prompt/code/pre block:

1. Confirm prompt text is not converted into live page structure.
2. Confirm angle brackets are escaped when needed.
3. Confirm configuration lines and exact-output templates retain intentional newlines.
4. Confirm prose-heavy prompt bodies do not contain arbitrary hard line breaks from extraction.
5. Confirm no Markdown fences wrap the whole artifact.
6. Confirm no UI scaffolding from Google Docs / Gemini / ChatGPT remains unless intentionally preserved as content.

AUDIT PHASE 6 — LINK AND MEDIA PARITY
When source is available:

- count source href values;
- count output href values;
- list missing hrefs;
- list added hrefs;
- check whether URLs were translated, split, or entity-damaged;
- count image/media src values;
- verify embeds remain in the expected relative positions;
- verify internal anchors still point to existing ids.

AUDIT PHASE 7 — METADATA / LANGUAGE / TRANSLATION FLAGS
Check:

- lang attributes;
- translate attributes;
- title/metadata remnants from a wrong language;
- duplicate ids;
- stale source-language labels in a translated page;
- accidental translation of ids/classes/hrefs/src.

AUDIT PHASE 8 — CONTENT SAFETY / DOCTRINAL NON-DRIFT
For ATR/Dharma content, flag:

- doctrinal upgrades not present in source;
- nondual/Buddhist terms reified by formatting/editing changes;
- quote boundaries lost;
- source labels removed;
- original-script quotations altered;
- Shared Termbank applied globally where context split is required;
- Prompt X source restorations made without confidence labels.

AUDIT PHASE 9 — CHANGELOG AND HANDOFF CONTINUITY
Check:

- latest batch changelog exists;
- changelog accurately describes actual file changes;
- no claim says a component was modernized if the body is absent;
- updated handoff prompt names the correct latest files;
- next batch recommendation is specific and honest.


AUDIT PHASE 10 — COSMIC DAWN v5.8.1 VISUAL / BLOGGER INTEGRATION
Check, where applicable:

- correct mode: resource-hub / teaching-article / article-showcase;
- teaching-article root has white outer canvas and no yellow gradient;
- no outer showcase shell narrows live Blogger teaching articles;
- stage cards fill available post width;
- orange stage-title bar appears in stage cards and does not appear in resource headings;
- mathematical purple→copper pill style is the final winning style for .stage-label / .guide-label;
- normal links are #2563eb on light backgrounds;
- dark boxes retain gold/cream links and emphasis;
- Available Translations headings are dark navy/slate, not brown/copper/gold;
- black callouts wrap only small isolated blocks;
- black-box strong/bold text remains high contrast;
- Blogger video macro exception is preserved.

COMBINED HISTORY / PATCH SUMMARY

Batch 60 / v5.8.5 adds:
- title contrast final-winning rule for purple/dark hero banners;
- final CSS order readback for .atr-hero h1;
- fail if the last applicable hero h1 color rule is not near-white/light cream.

Batch 57 / v5.8.3 incorporated v5.8.2 actual-DOM correction:
- no heavy top intro card;
- orange h2 bar must match actual h2.atr-section-title DOM;
- language/translation list uses a subtle soft panel;
- audit v1.9 checks selector-to-DOM applicability.

Batch 58 / v5.8.4 adds:
- purple/dark panel title contrast lock;
- dark-panel heading CSS must beat ordinary teaching-article h2/h3 rules;
- audit v1.9.1 checks dark-panel heading contrast separately from body text.

Core principle:
CSS-present is insufficient. DOM-applied and final-winning CSS must pass.

OUTPUT FORMAT
Return a concise but complete QA report with these sections:

1. Status: PASS / PASS WITH NOTES / NEEDS REVIEW / FAIL.
2. Files inspected.
3. Component inventory.
4. Critical issues.
5. Non-critical issues.
6. Old-vs-current preservation findings.
7. HTML structure findings.
8. Link/media findings.
9. Prompt-body escaping / line-wrap findings.
10. Changelog/handoff findings.
11. Required fixes before publication.
12. Cosmic Dawn v5.8.1 visual / Blogger integration findings.
13. Actual DOM / selector-match findings.
14. Dark / purple panel heading contrast findings.
15. Dropcap aesthetic findings.
16. Recommended next batch.

If producing a repaired artifact, also return:

- full HTML file;
- TXT mirror;
- QA/change-log report;
- updated handoff prompt.

FINAL RULE
Do not say “passed,” “complete,” “Blogger-ready,” or “no missing details” unless you actually inspected the saved artifact or the supplied exact current file.