AI turns documents or topics into real, native PowerPoint decks—with native shapes, transitions and animations, data-backed charts and tables on demand, audio narration from speaker notes, and support for your own .pptx templates. · by Hugo He
feat(strategist): free the three recommended directions from catalog framing
hugohe3d07790b32554
diff --git a/docs/technical-design.md b/docs/technical-design.mdindex 8bbd8b7f..2c0b3930 100644--- a/docs/technical-design.md+++ b/docs/technical-design.md@@ -495,7 +495,7 @@ without interaction, while the Stage-1 communication recommendation is authoredwithout reading those candidates or any template content. The UI confirms thatcommunication contract and the switchable free-design/template choice together;only afterward does the agent install/fuse a non-free selection. Strategist uses-one dependency-ordered two-stage gate. Its `delivery_context` recommendation distinguishes presenter-led, reader-led, hybrid, and recorded/self-running use in one prose field, names the primary context, and records any secondary use; hybrid never stands alone without its lead mode. The prose boxes remain editable and none requires a non-empty answer: confirmation persists the current text exactly, and a cleared value stays empty instead of falling back to the recommendation. Final Stage 2 is authored once from that contract. Strategist first creates exactly three complete, viable, project-fit solutions, then projects each into a custom mode, custom visual style, color, typography, icon, and custom generated-image rendering; the fixed catalogs remain conservative lower-level single-select alternatives. It does not assemble the three solutions afterward from independent component picks or force them into safe / shifted / bold archetypes. After all three bundles are complete, it compares them against the confirmed contract and source, records the strongest overall fit as zero-based `design_directions.selected`, and gives array order no preference meaning. Before reference selection it reads only the three catalog indexes; after each solution's bases are frozen, it reads only the deduplicated selected detail files and completes the custom behaviors. Image source remains an independent judgment, so every solution already has a rendering candidate even when AI imagery is initially not recommended. The preferred top card is marked and applied initially. An inactive top card applies its immutable bundle, each projected custom card becomes editable in place after selection, and all lower controls may diverge; the adjusted active card exposes an explicit restore action, while final state stores only the edited current components. The same stage confirms reading mode, page count, conditional AI acquisition, generation mode, the Design Spec review toggle, and whether the agent should proactively generate speaker notes, custom animations, or narration audio. Formula handling is not a user-selectable production policy: structural mathematics becomes a native formula marker during authoring. Only after Stage 1 is confirmed, when a template is installed, does Strategist read the project-local workspace and current content, derive **how to apply it**, and expose that plan as editable `template_application` prose; Stage 2 never reselects a template, and the internal reuse/adherence modes stay hidden. Reading mode decides how meaning is carried by page, visuals, presenter, and notes; its cards do not present px values. The browser may apply the deterministic `reading mode → body baseline → unpinned role sizes` dependency locally, while manual size edits pin visible values. It never regenerates Stage 2. The three proactive values are fallback policies, not capability bans: the latest explicit user instruction wins, then final Stage 2, then the fixed defaults `true / false / false`. Strategist still records non-binding Motion suggestions when useful; a suggestion alone does not activate custom-animation execution. Generated images inherit the selected deck color anchors directly; there is no independent image-palette choice. Final state has two equivalent carriers: the default UI path reads `confirm_ui/result.json` exactly once after the final wait, while an explicit chat-only or delegated path retains an equivalent final confirmation summary and may produce no `result.json`. Both paths first resolve and persist every effective production outcome into one `design_spec.md` and complete Gate 1 fidelity. With refinement off, lock authoring proceeds immediately. With `refine_spec: true`, the pipeline stops before `spec_lock.md`; the user may revise any part of that same Design Spec through normal chat for any number of rounds, and explicit approval then releases Gate 2 lock authoring. No second Design Spec or parallel lock is maintained. Normal lock authoring and downstream execution do not reread either confirmation channel. Required manual assets may still introduce their own conditional blocking points, so this is not an exclusive claim over every runtime gate. Project validation requires the compact `audience` / `objective` / `core_message` anchor set under `spec_lock.md ## communication` plus an `Audience move` in every §IX Slide block.+one dependency-ordered two-stage gate. Its `delivery_context` recommendation distinguishes presenter-led, reader-led, hybrid, and recorded/self-running use in one prose field, names the primary context, and records any secondary use; hybrid never stands alone without its lead mode. The prose boxes remain editable and none requires a non-empty answer: confirmation persists the current text exactly, and a cleared value stays empty instead of falling back to the recommendation. Final Stage 2 is authored once from that contract. Strategist first creates exactly three complete, viable, project-fit solutions, then projects each into a custom mode, custom visual style, color, typography, icon, and custom generated-image rendering; the fixed catalogs remain conservative lower-level single-select alternatives. Each direction is one complete design authored top-down within the confirmed contract rather than assembled from catalog picks, and three exist so a single recommendation cannot lock the user in; the fixed catalogs stay the manual lower layer. `custom` has no required relationship to the catalogs, and one selected preset may be carried unchanged. The three bundles are plainly different designs at the whole-deck level, decided before any field is written and only then projected onto mode, visual style, rendering, color, typography, and icons; that difference lives in the solutions rather than in one designated field, so whichever components a direction requires carry it and any component may coincide. A different name, note, or reference count alone is not a difference, and bundles identical on every component are not three solutions. Where authoritative user/template truth fixes some components, the remaining open ones carry the difference; identical projections are valid only when nothing is open, and the plan states that boundary instead of inventing variation. It does not assemble the three solutions afterward from independent component picks or force them into safe / shifted / bold archetypes. After all three bundles are complete, it compares them against the confirmed contract and source, records the strongest overall fit as zero-based `design_directions.selected`, and gives array order no preference meaning. Before reference selection it reads only the three catalog indexes; after each solution's bases are frozen, it reads only the deduplicated selected detail files and completes the custom behaviors. Image source remains an independent judgment, so every solution already has a rendering candidate even when AI imagery is initially not recommended. The preferred top card is marked and applied initially. An inactive top card applies its immutable bundle, each projected custom card becomes editable in place after selection, and all lower controls may diverge; the adjusted active card exposes an explicit restore action, while final state stores only the edited current components. The same stage confirms reading mode, page count, conditional AI acquisition, generation mode, the Design Spec review toggle, and whether the agent should proactively generate speaker notes, custom animations, or narration audio. Formula handling is not a user-selectable production policy: structural mathematics becomes a native formula marker during authoring. Only after Stage 1 is confirmed, when a template is installed, does Strategist read the project-local workspace and current content, derive **how to apply it**, and expose that plan as editable `template_application` prose; Stage 2 never reselects a template, and the internal reuse/adherence modes stay hidden. Reading mode decides how meaning is carried by page, visuals, presenter, and notes; its cards do not present px values. The browser may apply the deterministic `reading mode → body baseline → unpinned role sizes` dependency locally, while manual size edits pin visible values. It never regenerates Stage 2. The three proactive values are fallback policies, not capability bans: the latest explicit user instruction wins, then final Stage 2, then the fixed defaults `true / false / false`. Strategist still records non-binding Motion suggestions when useful; a suggestion alone does not activate custom-animation execution. Generated images inherit the selected deck color anchors directly; there is no independent image-palette choice. Final state has two equivalent carriers: the default UI path reads `confirm_ui/result.json` exactly once after the final wait, while an explicit chat-only or delegated path retains an equivalent final confirmation summary and may produce no `result.json`. Both paths first resolve and persist every effective production outcome into one `design_spec.md` and complete Gate 1 fidelity. With refinement off, lock authoring proceeds immediately. With `refine_spec: true`, the pipeline stops before `spec_lock.md`; the user may revise any part of that same Design Spec through normal chat for any number of rounds, and explicit approval then releases Gate 2 lock authoring. No second Design Spec or parallel lock is maintained. Normal lock authoring and downstream execution do not reread either confirmation channel. Required manual assets may still introduce their own conditional blocking points, so this is not an exclusive claim over every runtime gate. Project validation requires the compact `audience` / `objective` / `core_message` anchor set under `spec_lock.md ## communication` plus an `Audience move` in every §IX Slide block.The three proactive values confirmed in final Stage 2 remain independent rawevidence. In particular, enabling narration may enable the effective Speaker@@ -574,7 +574,7 @@ Why both? `design_spec.md` preserves the complete confirmed solution and rationaThe view omits universal SVG/icon prohibitions already owned by the always-loaded Executor core and retains only project-specific forbidden rows. It selects images from the current page brief, explicit image-resource page assignments, and mirror prototype references. Images assigned elsewhere are excluded; any unresolved legacy image remains in a compatibility subset, and `confirmed-none` appears only when every locked image has a deterministic assignment elsewhere.-The lock is also the per-page routing table. Beyond global colors and typography, it carries `page_rhythm` (`anchor` / `dense` / `breathing`), `page_visualizations` (the selected page-local `chart|table/<key>` reference, which triggers reading the corresponding family SVG and §VII Usage but does not lock the final information model, type, or geometry), image rows with placement/cropping contracts, and the locked `mode` / `visual_style` values. Each decision owner first reads its complete compact capability map: Chart planning reads the objective 33-entry expression vocabulary rather than authoring instructions, Table planning reads its small registry, and Executor reads the full 187-name preset-shape vocabulary before local contour selection. After a reference is selected, only its exact detail or implementation source enters the owning role context. Qualitative Structure is described directly in §IX and generated from those relationships; it has no §VII row, `page_visualizations` key, or catalog SVG. Legacy `page_charts` remains readable but is not written by new planning; a historical Structure bare key is retained only as semantic intent without an SVG reference. A selected custom direction also carries its resolved `mode_behavior` / `visual_style_behavior`. It may use zero, one, or many catalog bases; when any are used, the lock records their exact ids in optional `mode_references` / `visual_style_references` / `image_rendering_references`. Reference count has no fixed cap, but every listed id must contribute a distinct executable job and no second id is required for a one-basis specialization. Those ids preserve provenance, constrain synthesis to the declared bases, and route the only permitted detail-file reads. A genuinely novel custom omits the corresponding reference field. `template_reuse_scope: mirror|layout` locks additionally carry `page_layouts` (which input template SVG each page inherits), unique `pptx_masters` / `pptx_layouts` definitions, and `page_pptx_layouts` assignments. `template_reuse_scope: style`, free-design, and brand-only projects use `pptx_structure.mode: flat` and omit those sections entirely rather than writing empty values. Empty entries elsewhere are meaningful signals: no catalog reference or no image is often a design decision rather than missing data.+The lock is also the per-page routing table. Beyond global colors and typography, it carries `page_rhythm` (`anchor` / `dense` / `breathing`), `page_visualizations` (the selected page-local `chart|table/<key>` reference, which triggers reading the corresponding family SVG and §VII Usage but does not lock the final information model, type, or geometry), image rows with placement/cropping contracts, and the locked `mode` / `visual_style` values. Each decision owner first reads its complete compact capability map: Chart planning reads the objective 33-entry expression vocabulary rather than authoring instructions, Table planning reads its small registry, and Executor reads the full 187-name preset-shape vocabulary before local contour selection. After a reference is selected, only its exact detail or implementation source enters the owning role context. Qualitative Structure is described directly in §IX and generated from those relationships; it has no §VII row, `page_visualizations` key, or catalog SVG. Legacy `page_charts` remains readable but is not written by new planning; a historical Structure bare key is retained only as semantic intent without an SVG reference. A selected custom direction also carries its resolved `mode_behavior` / `visual_style_behavior`. It may use zero, one, or many catalog bases; when any are used, the lock records their exact ids in optional `mode_references` / `visual_style_references` / `image_rendering_references`. Reference count has no fixed cap. One listed id may carry its complete preset behavior unchanged; with several, each contributes a distinct executable job. Those ids preserve provenance, constrain realization to the declared bases, and route the only permitted detail-file reads. A genuinely novel custom omits the corresponding reference field. `template_reuse_scope: mirror|layout` locks additionally carry `page_layouts` (which input template SVG each page inherits), unique `pptx_masters` / `pptx_layouts` definitions, and `page_pptx_layouts` assignments. `template_reuse_scope: style`, free-design, and brand-only projects use `pptx_structure.mode: flat` and omit those sections entirely rather than writing empty values. Empty entries elsewhere are meaningful signals: no catalog reference or no image is often a design decision rather than missing data.`page-context v2` remains an on-demand projector. Each invocation emits a compact global anchor set bound to `lock_source.sha256`, the current §IX/resource/routing delta, and scoped path/SHA fingerprints for large references; the projection is not a color/font allowlist or a replacement authority. It is used only for explicit diagnostics/telemetry or an unresolved page/template/visualization path-SHA question. A bounded exact repair in a valid uncompacted context may use targeted fragment readback plus validation; fresh, compacted, external, unknown, structural, or mismatched changes require one complete Design Spec and lock read. Flat pages have no prototype reference; structured pages use the authoritative complete SVG. Manifests and text-slot sidecars remain derived tool diagnostics and are never injected into page authoring.@@ -647,7 +647,7 @@ Several architectural decisions shape this phase:**External refs during development, two divergent embedding strategies for downstream outputs.** While editing in `svg_output/`, images are external file references for fast iteration and single-point replacement. `svg_final/` then Base64-inlines ordinary bitmaps and supported SVG resources, while retaining external EMF/WMF references for native PPTX passthrough; an individual inline failure increments the processing-error count, but it does not rewrite authored source or fail the finalizer as a whole. Native PPTX instead copies bitmaps or Office vectors into the package media folder and uses `<a:srcRect>` for supported bitmap cropping. The split exists because the preview and compiler serve different responsibilities: one is a derived visual reference, while the other produces editable DrawingML through the project converter. `svg_final/` is not an unconditionally portable no-external-dependency interchange format or a compatibility source for PowerPoint Convert to Shape.-**One rendering lock, inherited deck anchors, and per-image composition.** Before deciding whether to recommend AI imagery, Stage 2 authors one complete custom deck-wide rendering candidate inside each of its three coordinated design directions. It uses the rendering index to freeze any bases before reading only those selected detail files. The source decision is separate: the UI hides rendering controls until AI is selected, then reveals the three editable custom candidates followed by the fixed rendering catalog without another recommendation pass. Only confirmed AI usage promotes the selected current value into the execution contract. Image colors are not a second user decision: Image_Generator starts from the core HEX roles in `spec_lock.md colors`, reads only the selected rendering entry or exact custom bases, then interprets them with the completed Design Spec and each resource's `type` or hero-page composition. Rendering may derive coherent tints, material colors, lighting transitions, and atmospheric hues while preserving the core role meanings; it must not replace the deck identity with an unrelated image-only palette. A repeated derived tone may be promoted into a named lock role.+**One rendering lock, inherited deck anchors, and per-image composition.** Before deciding whether to recommend AI imagery, Stage 2 authors one complete custom deck-wide rendering candidate inside each of its three coordinated design directions. It uses the rendering index to freeze any bases before reading only those selected detail files; one selected rendering may supply the full treatment unchanged without a fabricated delta. The source decision is separate: the UI hides rendering controls until AI is selected, then reveals the three editable custom candidates followed by the fixed rendering catalog without another recommendation pass. Only confirmed AI usage promotes the selected current value into the execution contract. Image colors are not a second user decision: Image_Generator starts from the core HEX roles in `spec_lock.md colors`, reads only the selected rendering entry or exact custom bases, then interprets them with the completed Design Spec and each resource's `type` or hero-page composition. Rendering may derive coherent tints, material colors, lighting transitions, and atmospheric hues while preserving the core role meanings; it must not replace the deck identity with an unrelated image-only palette. A repeated derived tone may be promoted into a named lock role.---diff --git a/docs/zh/technical-design.md b/docs/zh/technical-design.mdindex 88d7c9a0..d6300b44 100644--- a/docs/zh/technical-design.md+++ b/docs/zh/technical-design.md@@ -420,6 +420,8 @@ Stage-1 沟通推荐的编写不得读取这些候选或任何模板内容。UI确认沟通契约与可切换的自由设计/使用模板选择;只有确认后,Agent 才安装或合成非自由设计选择。Strategist 采用一个按依赖排序的两阶段 gate。`delivery_context` 在同一个开放文本字段中区分演讲者主导、读者主导、混合、录制/自动播放,明确主要场景并记录可选的次要用途;混合场景不能只写“混合”而不说明由哪一种主导。其中的文本框仍承载可编辑推荐,且没有任何一项要求非空:确认时按当前文本原样保存,清空后的值保持为空,不会回退到推荐内容。最终 Stage 2 只从该契约计算一次,同时确认完整 PPT 方案与生产机制:阅读模式、叙事 mode、页数、成套视觉系统、图片来源、生成图渲染、条件式 AI 图片获取路径、生成模式、Design Spec 审核开关,以及 Agent 是否主动生成讲稿、自定义动画和旁白音频。公式处理不是用户选择的生产策略:结构性数学内容在创作时直接成为原生公式 marker。仅在 Stage 1 确认后,存在已安装模板时,Strategist 才读取项目本地工作区和当前内容,推导**如何应用**并以可编辑的 `template_application` 自然语言展示;Stage 2 不会重新选模板,内部复用/遵循模式保持隐藏。阅读模式决定信息由页面、视觉、讲者和备注如何共同承担,其选项卡不展示 px 数值。浏览器可以在本地执行确定性的「阅读模式 → 正文基准 → 未锁定角色字号」联动;手动编辑字号即锁定可见值,不会重新计算 Stage 2。三个主动执行值是用户未明确要求时的兜底策略,不是能力禁用开关;优先级固定为「用户最新明确指令 → 最终 Stage 2 → `true / false / false` 默认值」。Strategist 仍可按内容给出非绑定的 Motion suggestion,建议本身不会启动自定义动画执行。生成图直接继承已选 PPT 色彩锚点,不再单设图片调色选择。最终状态有两个等价载体:默认 UI 路径在最终等待返回后只读取一次 `confirm_ui/result.json`;显式 chat-only 或委托路径保留等价的最终确认摘要,并可不产生 `result.json`。两条路径都会先解析并把生产流程的最终有效结果固化到同一份 `design_spec.md`,再完成 Gate 1 fidelity。未开启 refine 时立即进入 lock 编写;`refine_spec: true` 时,流程会在 `spec_lock.md` 之前暂停,用户可以通过正常聊天任意修改这同一份 Design Spec、迭代任意轮次,明确批准后才释放 Gate 2 并编写 lock。流程不会维护第二份 Design Spec 或并行 lock。正常的 lock 编写与下游执行不再回读确认通道。必需人工素材未就绪仍可能引入条件式阻塞点,因此这里不是对所有 runtime gate 的排他声明。项目校验要求 `spec_lock.md ## communication` 下存在紧凑的 `audience` / `objective` / `core_message` 锚点,并要求 §IX 每个 Slide block 都有 `Audience move`。+三套成套方向仍把 mode、visual style 与生成图 rendering 投影为可编辑的 `custom`。每套方向都是在已确认契约内自顶向下完成的一套完整设计,而不是自下而上从目录里拼选出来的;给三套是为了不让单一推荐把用户锁死,固定 preset 目录始终是供用户手动选择的下层。其 custom 投影不受“与 preset 是什么关系”的约束,可以原样承接一个 preset。`custom` 与 preset 目录没有固定关系,一个最合适的 preset 也可以原样成为完整基础。三套方案在整份 deck 的层面本身就显著不同:先定下三个大方案,再补全各自细节,最后才映射到 mode、visual style、rendering、色彩、字体和图标等具体选项。差异存在于方案本身,而不是由某个指定字段来证明——由该方案设计真正需要的组件承担,任何组件都可以承担,也任何组件都允许一致。仅换名称、note 或参考数量不构成差异;所有组件都完全相同的三项也不成其为三套方案。当用户或模板的真值锁定了部分组件时,由其余开放组件承担差异;只有在没有任何开放组件时投影才可以完全相同,并须如实说明边界而不是强行制造差异。+最终 Stage 2 确认的三个主动执行值始终保留为相互独立的原始证据。尤其是,启用旁白可以在 Design Spec 中启用 Speaker Notes 的最终有效结果,但绝不会改写原始的备注选择。@@ -484,7 +486,7 @@ Generate 路由会在加载流程前选定一份运行时权威:[`workflows/ge该视图省略 Executor core 已经恒定加载的通用 SVG/图标禁令,只保留项目专属 forbidden 行。图片从当前页 brief、图片资源表中的显式页分配和 mirror 原型引用中选择;已分配给其他页面的图片会被排除,仍无法归属的 legacy 图片保留在兼容子集中,只有所有锁定图片都能确定归属到其他页面时才记录 `confirmed-none`。-这份 lock 同时也是逐页路由表。除了全局配色和字体,它还承载 `page_rhythm`(`anchor` / `dense` / `breathing`)、`page_visualizations`(某页选中的 `chart|table/<key>` 页面级参考;它只触发读取对应 family SVG 与 §VII Usage,不锁定最终信息模型、类型或几何)、带放置/裁剪契约的图片行,以及决定加载哪些执行规则文件的 `mode` / `visual_style`。每个决策所有者会先读取自己完整而紧凑的能力地图:Chart 规划读取客观的 33 项表达词汇表,而不是制作说明;Table 规划读取其小型 registry;Executor 在局部轮廓选择前读取完整 187 项 preset shape 词汇表。选中参考后,只有对应的精确细节或实现来源进入职责上下文。定性 Structure 直接在 §IX 描述并根据这些关系现场生成,没有 §VII 行、`page_visualizations` key 或 catalog SVG。旧 `page_charts` 仍可读取,但新规划不再写入;历史 Structure bare key 只保留为不带 SVG reference 的语义意图。选定 custom 方向时,lock 还承载已消解的 `mode_behavior` / `visual_style_behavior`;当它确实综合或借鉴已有目录项时,再用可选的 `mode_references` / `visual_style_references` / `image_rendering_references` 记录全部精确 id,执行阶段会先读取每个对应文件再综合。真正全新的 custom 不写参考字段。`template_reuse_scope: mirror|layout` 项目的 lock 还承载 `page_layouts`(每页继承哪个输入模板 SVG)、唯一的 `pptx_masters` / `pptx_layouts` 定义,以及 `page_pptx_layouts` 页面分配;`template_reuse_scope: style`、自由设计和 brand-only 项目使用 `pptx_structure.mode: flat`,那些段整段省略,而不是写成空值。其余字段的空值本身仍是信号:没有 catalog 参考、没有图片,很多时候是设计选择,而不是漏填。+这份 lock 同时也是逐页路由表。除了全局配色和字体,它还承载 `page_rhythm`(`anchor` / `dense` / `breathing`)、`page_visualizations`(某页选中的 `chart|table/<key>` 页面级参考;它只触发读取对应 family SVG 与 §VII Usage,不锁定最终信息模型、类型或几何)、带放置/裁剪契约的图片行,以及决定加载哪些执行规则文件的 `mode` / `visual_style`。每个决策所有者会先读取自己完整而紧凑的能力地图:Chart 规划读取客观的 33 项表达词汇表,而不是制作说明;Table 规划读取其小型 registry;Executor 在局部轮廓选择前读取完整 187 项 preset shape 词汇表。选中参考后,只有对应的精确细节或实现来源进入职责上下文。定性 Structure 直接在 §IX 描述并根据这些关系现场生成,没有 §VII 行、`page_visualizations` key 或 catalog SVG。旧 `page_charts` 仍可读取,但新规划不再写入;历史 Structure bare key 只保留为不带 SVG reference 的语义意图。选定 custom 方向时,lock 还承载已消解的 `mode_behavior` / `visual_style_behavior`;当它使用已有目录项时,再用可选的 `mode_references` / `visual_style_references` / `image_rendering_references` 记录全部精确 id。执行阶段对单个参考按 behavior 直接应用;只有多个参考才按各自贡献综合。未使用目录项的 custom 不写参考字段。`template_reuse_scope: mirror|layout` 项目的 lock 还承载 `page_layouts`(每页继承哪个输入模板 SVG)、唯一的 `pptx_masters` / `pptx_layouts` 定义,以及 `page_pptx_layouts` 页面分配;`template_reuse_scope: style`、自由设计和 brand-only 项目使用 `pptx_structure.mode: flat`,那些段整段省略,而不是写成空值。其余字段的空值本身仍是信号:没有 catalog 参考、没有图片,很多时候是设计选择,而不是漏填。`page-context v2` 保留为按需投影器。每次调用都会输出绑定 `lock_source.sha256` 的紧凑全局锚点、当前 §IX/资源/路由 delta,以及大型引用的带 scope 路径/SHA 指纹;该投影既不是颜色/字体白名单,也不是替代权威。只有明确的诊断/统计需求,或页面/模板/可视化参考的路径-SHA 问题仍未解决时才调用。有效且未压缩的上下文中做有界精确回修,可以只回读修改片段并校验;fresh、已压缩、外部、来源不明、结构性或投影不符的修改必须完整读取 Design Spec 与 lock。flat 页面没有原型引用;structured 页面使用权威完整 SVG。manifest 与 text-slot sidecar 只保留为派生工具诊断,不注入页面创作上下文。@@ -543,7 +545,7 @@ Generate 路由会在加载流程前选定一份运行时权威:[`workflows/ge**开发期外部引用,下游分叉成预览与原生导出两套嵌入策略。** 在 `svg_output/` 里编辑时,图片是外部文件引用——快速迭代、单点替换。随后分成两种表达:`svg_final/` 对普通位图和受支持的 SVG 资源执行 Base64 内联,EMF/WMF 则保留外部引用供 native PPTX passthrough;个别资源内联失败会计入处理错误,但不会回写作者源或让 finalizer 整体失败。native PPTX 则把位图或 Office vector 复制进 PPTX 的 media 文件夹,用 `<a:srcRect>` 表达受支持的位图裁剪。分叉的理由是职责不同:前者服务视觉预览,后者服务项目转换器生成的可编辑 DrawingML。`svg_final/` 不是无条件脱离项目资产即可搬运的交换格式,也不作为 PowerPoint 手工“转换为形状”的兼容源。-**一份渲染锁、继承 PPT 色彩锚点、逐图确定构图。** 当 deck 包含 AI 生成图片时,Stage 2 会在每套成套设计方向中确认 deck 级 `rendering`。图片颜色不再形成第二次用户决策:Image_Generator 从 `spec_lock.md colors` 的核心 HEX 角色出发,再结合完整 Design Spec 与每项资源按用途推导的 `type` 或 hero-page 构图。渲染可以在不改变核心角色语义的前提下,按上下文派生色阶、材质色、明暗过渡与氛围色;不得用一套无关的图片专属调色替换 deck 身份。重复使用的派生色可以提升为具名 lock 角色。+**一份渲染锁、继承 PPT 色彩锚点、逐图确定构图。** 当 deck 包含 AI 生成图片时,Stage 2 会在每套成套设计方向中确认 deck 级 `rendering`;一个选定的 rendering preset 可以不加人为差异地提供完整处理。图片颜色不再形成第二次用户决策:Image_Generator 从 `spec_lock.md colors` 的核心 HEX 角色出发,再结合完整 Design Spec 与每项资源按用途推导的 `type` 或 hero-page 构图。渲染可以在不改变核心角色语义的前提下,按上下文派生色阶、材质色、明暗过渡与氛围色;不得用一套无关的图片专属调色替换 deck 身份。重复使用的派生色可以提升为具名 lock 角色。---diff --git a/skills/ppt-master/references/executor-base.md b/skills/ppt-master/references/executor-base.mdindex 1c6dd2dc..ba1466ad 100644--- a/skills/ppt-master/references/executor-base.md+++ b/skills/ppt-master/references/executor-base.md@@ -19,7 +19,7 @@ Always-loaded Executor authority for flat SVG page authoring and behavior sharedEvaluate branches from each object's actual information model, not only from a Chart/Table reference. A catalog family selects construction guidance but never native readiness; `Native-ready` is an independent object-level decision. Page-local qualitative geometry also never implies package-level `pptx_structure.mode: structured`.-> Narrative skeleton and visual aesthetic come from the locked values selected through the [`modes/`](./modes/_index.md) and [`visual-styles/`](./visual-styles/_index.md) planning indexes. Executor does not reopen those indexes: it reads one locked preset file or only the exact `*_references` of a custom synthesis; an unreferenced novel custom reads none. [`shared-standards-core.md`](./shared-standards-core.md) supplies the technical boundary plus the fallback visual-quality and leading defaults when those authorities are silent.+> Narrative skeleton and visual aesthetic come from the locked values selected through the [`modes/`](./modes/_index.md) and [`visual-styles/`](./visual-styles/_index.md) planning indexes. Executor does not reopen those indexes: it reads one locked preset file or only the exact `*_references` of a custom, applying one basis under its behavior or synthesizing several by their stated contributions; an unreferenced custom reads none. [`shared-standards-core.md`](./shared-standards-core.md) supplies the technical boundary plus the fallback visual-quality and leading defaults when those authorities are silent.**Hard rule — Shape-first page authority**: Every visible object intended for the exported slide MUST exist in the final page SVG or be explicitly referenced by it. Templates and `spec_lock.md` guide construction; they are not export-time overlays for missing visible content. Optional native Chart/Table metadata belongs to an independently selected object and never replaces this visible fallback during authoring; [`native-data-interface.md`](./native-data-interface.md) alone defines that metadata and its export activation. Native formula markers require a matching SVG preview; export replaces only it under [`native-formula.md`](./native-formula.md).diff --git a/skills/ppt-master/references/image-generator.md b/skills/ppt-master/references/image-generator.mdindex 442f7d7b..3259fd10 100644--- a/skills/ppt-master/references/image-generator.md+++ b/skills/ppt-master/references/image-generator.md@@ -93,7 +93,7 @@ Use them as identity anchors. Do not create another user-facing image-color choi**Quick Generate path**: the main agent resolves one active-context rendering/color set, honoring explicit user values and deciding the rest without interaction. Write it to `image_prompts.json`; create no planning artifacts.-**Hard rule — `custom` catalog basis**: when `image_rendering` is `custom`, first inspect the optional `image_rendering_references` row. If present, read every exact `image-renderings/<id>.md` it lists and synthesize their line, texture, depth, material, and mood guidance under `image_rendering_behavior` before assembling prompts. If absent, the custom is genuinely novel: read no preset file and use `image_rendering_behavior` directly. Never infer or add adjacent references during execution. The deck color-role rows remain authoritative.+**Hard rule — `custom` catalog basis**: when `image_rendering` is `custom`, first inspect the optional `image_rendering_references` row. If present, read every exact `image-renderings/<id>.md` it lists. Apply one basis under `image_rendering_behavior`, unchanged when that behavior carries it as-is; synthesize several only by their stated line, texture, depth, material, and mood contributions. If absent, read no preset file and use `image_rendering_behavior` directly. Never infer or add adjacent references during execution. The deck color-role rows remain authoritative.**Declared-inference fallback — when an existing `spec_lock.md` omits the `image_rendering` key** (see [`failure-recovery.md`](../workflows/governance/failure-recovery.md) §2):diff --git a/skills/ppt-master/references/image-renderings/_index.md b/skills/ppt-master/references/image-renderings/_index.mdindex db2ab9c9..25cdaf89 100644--- a/skills/ppt-master/references/image-renderings/_index.md+++ b/skills/ppt-master/references/image-renderings/_index.md@@ -52,9 +52,9 @@ Each rendering keeps its own authoritative file with: style paragraph, line / te| [`pixel-art`](./pixel-art.md) | 8-bit retro game aesthetic | Gaming / retro tech / nostalgic || [`nature`](./nature.md) | Organic earthy illustration | Environment / wellness / sustainability |-### 1.5 Escape hatch — `custom`+### 1.5 Editable `custom` projection-Every coordinated Stage-2 direction carries one complete `rendering: custom` candidate even when `recommend.image_usage` does not include `ai`. The UI keeps rendering controls hidden until the current source selection includes AI, then exposes the three already-authored project candidates without another backend recommendation. The 20 fixed renderings remain lower-level single-select alternatives. A template-backed proposal must honor inherited identity and the confirmed template-application plan.+Every coordinated Stage-2 direction carries one complete `rendering: custom` candidate even when `recommend.image_usage` does not include `ai`. The UI keeps rendering controls hidden until the current source selection includes AI, then exposes the three already-authored project candidates without another backend recommendation. `custom` is not constrained by its relationship to the catalog: it may use catalog material in any way or none, including carrying one fitting preset treatment unchanged. The three complete directions are plainly different designs, but no single component is required to carry that difference: rendering treatments and bases may coincide when other components express it, while a different name, note, or reference count alone is never a difference. The 20 fixed renderings remain lower-level single-select alternatives. A template-backed proposal must honor inherited identity and the confirmed template-application plan.**Hard rule — `rendering_behavior` prose**:@@ -62,7 +62,7 @@ Every coordinated Stage-2 direction carries one complete `rendering: custom` can|---|---|| Length | One paragraph, 2-5 sentences || Axes covered | line / texture / depth / material / mood (same as preset files) |-| Catalog basis | Freeze every exact id from this index, then read only those named files before synthesis |+| Catalog basis | Freeze every exact id from this index, then read only those files before applying one basis or synthesizing several |```yaml- image_rendering: custom@@ -71,7 +71,7 @@ Every coordinated Stage-2 direction carries one complete `rendering: custom` can**Hard rule**: three complete rendering candidates are mandatory in every fresh Stage-2 direction set; AI source recommendation remains independent. See [`strategist-image.md`](../strategist-image.md) for the Stage-2 carrier and downstream lock behavior.-Write `image_rendering_references` only when the confirmed custom direction actually uses catalog material. A custom may use zero, one, or many renderings: keep one when it owns the whole specialized treatment, or include every rendering that contributes a distinct executable job across line, texture, depth, material, or mood. Reference count has no fixed cap; count is an outcome, not a target. A four-basis direction may assign `vector-illustration` to silhouette clarity, `minimalist-swiss` to negative-space composition, `screen-print` to restrained halftone texture, and `warm-scene` to light and mood; list all four ids. Omit every rendering whose contribution cannot be stated and never add a second merely to imply synthesis. A genuinely new rendering with no catalog source omits the field and proceeds from its standalone behavior; never invent a reference merely to legitimize `custom`.+Write `image_rendering_references` only when the confirmed custom direction actually uses catalog material. One rendering may supply the complete treatment unchanged; when several are named, each contributes a distinct executable job across line, texture, depth, material, or mood. Reference count has no fixed cap; count is an outcome, not a target. A four-basis direction may assign `vector-illustration` to silhouette clarity, `minimalist-swiss` to negative-space composition, `screen-print` to restrained halftone texture, and `warm-scene` to light and mood; list all four ids. Omit every rendering whose contribution cannot be stated and never add a second merely to imply synthesis. A custom using no catalog source omits the field and proceeds from its standalone behavior; never invent a reference merely to legitimize `custom`.---@@ -99,6 +99,6 @@ describes the intended treatment, use `custom` per §1.5.1. Read the resolved visual system and the deck's intended AI image jobs.2. Compare the complete catalog and choose the strongest whole-deck fit, or use a warranted `custom` treatment.-3. For a preset, read `image-renderings/<chosen>.md`. For `custom`, read every file named in `image_rendering_references`, then synthesize them under the confirmed behavior; with no references, use the novel behavior directly. Apply the result when assembling prompts per [`image-generator.md`](../image-generator.md) §4.+3. For a preset, read `image-renderings/<chosen>.md`. For `custom`, read every file named in `image_rendering_references`: apply one basis under the confirmed behavior, or synthesize several by their stated contributions. With no references, use the behavior directly. Apply the result when assembling prompts per [`image-generator.md`](../image-generator.md) §4.**Lock for the whole deck.** Don't change rendering between images in the same deck.diff --git a/skills/ppt-master/references/modes/_index.md b/skills/ppt-master/references/modes/_index.mdindex f6986863..79b40f69 100644--- a/skills/ppt-master/references/modes/_index.md+++ b/skills/ppt-master/references/modes/_index.md@@ -58,18 +58,18 @@ selecting one for the current deck:----## 4. Escape hatch — `custom`+## 4. Editable projection and escape hatch — `custom`-`custom` holds **any bespoke narrative direction the five don't give as-is** — and what *kind* of thing it is doesn't matter. It might be a nameable cadence (dialectic 正反合, myth-vs-reality, countdown / Top-N, Socratic), a deliberate multi-act fusion of several modes, or the user's own feel for how the deck should carry (confrontational here, detached there). Don't try to taxonomize it.+`custom` is the editable behavior carrier, not a category defined by how it relates to the catalog. Default uses it for each coordinated direction even when one preset supplies the complete cadence unchanged. In Quick or a lower-level manual choice it remains the escape hatch for **a bespoke narrative direction the five don't give as-is**: a nameable cadence (dialectic 正反合, myth-vs-reality, countdown / Top-N, Socratic), a deliberate multi-act fusion, or the user's own posture shifts. Don't try to taxonomize those bespoke cases.-**Default candidates**: All three coordinated Stage-2 directions use literal `custom` plus a visible, non-empty `mode_behavior`. A direction may specialize one preset, fuse several modes, or define a novel cadence; it fits any installed template capacity. The fixed five remain lower-level single-select alternatives. Strategist crystallizes the confirmed current value in the Design Spec first, then projects its behavior and actual catalog basis to `spec_lock.md`.+**Default candidates**: All three coordinated Stage-2 directions use literal `custom` plus a visible, non-empty `mode_behavior`. A custom may use catalog material in any way or none, including carrying one preset unchanged; it fits any installed template capacity. The three complete directions are plainly different designs, but no single component is required to carry that difference: mode behaviors and bases may coincide when other components express it, while a different name, note, or reference count alone is never a difference. The fixed five remain lower-level single-select alternatives. Strategist crystallizes the confirmed current value in the Design Spec first, then projects its behavior and actual catalog basis to `spec_lock.md`.**Quick custom**: do not display a candidate set. Use `custom` only when a project-specific specialization or fusion serves the deck better than one preset; retain the behavior and exact bases in active context and persist nothing.-**Mandatory — select before detail reading**: Use this index to freeze every catalog source actually used, then read only those exact files before writing the behavior. A custom may use zero, one, or many sources: keep one when it owns the whole specialized cadence, or include every mode that owns a distinct executable act, posture, title voice, rhythm, or register. Reference count has no fixed cap; count is an outcome, not a target. A three-basis direction may use `pyramid` for a conclusion-first opening, `narrative` for the risk-tension act, and `instructional` for the closing action sequence; it reads those three files and writes all three ids beside `mode_behavior`. Quick retains its bases only in active context. Omit every source whose contribution cannot be stated, never add a second merely to imply synthesis, and do not open candidates for comparison after this gate. A genuinely new cadence names and reads no catalog source.+**Mandatory — select before detail reading**: Use this index to freeze every catalog source actually used, then read only those exact files before writing the behavior. One source may supply the complete cadence unchanged; when several are named, each owns a distinct executable act, posture, title voice, rhythm, or register. Reference count has no fixed cap; count is an outcome, not a target. A three-basis direction may use `pyramid` for a conclusion-first opening, `narrative` for the risk-tension act, and `instructional` for the closing action sequence; it reads those three files and writes all three ids beside `mode_behavior`. Quick retains its bases only in active context. Omit every source whose contribution cannot be stated, never add a second merely to imply synthesis, and do not open candidates for comparison after this gate. A custom using no catalog source names and reads none.> **One value per deck — fusion is *one* `custom`, not several modes.** A deck always resolves a single `mode`. A multi-mode blend is expressed as **one** custom behavior whose paragraph describes the acts — never as several simultaneous modes.>-> **Custom need not mean fusion.** A Default recommendation can specialize one dominant preset for this project's act sequence, title voice, rhythm, and register. Quick or a lower-level manual choice may still use the fixed preset directly.+> **Default custom need not mean fusion or deviation.** It may project one dominant preset unchanged into an editable behavior, or specialize it when the project actually calls for a delta. Quick or a lower-level manual choice still uses the fixed preset directly when no editable projection is needed.-**Forbidden — empty customization**: Do not relabel an index row as `custom`. The behavior must state the project-specific cadence, posture, title voice, page rhythm, or act sequence that the fixed preset alone does not encode. A user-stated direction remains authoritative the same way a user-supplied outline is — see the lens-not-mandate note in §1.+**Forbidden — empty behavior**: Single-preset reuse is valid. When a Default direction carries an index row without changing its logic, name and read that exact basis, then state its executable cadence, posture, title voice, page rhythm, or act sequence in the editable behavior; do not add another mode merely to justify `custom`. A bare label with no executable behavior remains invalid. A user-stated direction remains authoritative the same way a user-supplied outline is — see the lens-not-mandate note in §1.diff --git a/skills/ppt-master/references/strategist-image.md b/skills/ppt-master/references/strategist-image.mdindex 40722b59..787f0c24 100644--- a/skills/ppt-master/references/strategist-image.md+++ b/skills/ppt-master/references/strategist-image.md@@ -63,9 +63,9 @@ When a subject crosses a native title, panel, frame, or shape, the prepared path## 2. AI Image Strategy — always propose three; lock only for confirmed `ai`-Before any rendering detail, use the already-loaded [`image-renderings/_index.md`](./image-renderings/_index.md) as the sole rendering-basis catalog authority. First author exactly three complete, project-fit solution intents; use the index to freeze each intent's exact rendering bases, then read once only the deduplicated referenced sibling files. Project one complete `image_strategy` into each direction regardless of `recommend.image_usage`. Every candidate carries localized `name`, `rendering: custom`, `visual`, `mood`, and non-empty localized `behavior`. Mood includes a recognizable real-world analogy. All three must credibly serve their owning whole solution, but they need not use different bases or span artificial safe / shifted / bold extremes. Image colors always inherit that direction's deck HEX roles; never add an image palette or alter deck colors to rescue a rendering.+Before any rendering detail, use the already-loaded [`image-renderings/_index.md`](./image-renderings/_index.md) as the sole rendering-basis catalog authority. First author exactly three complete, project-fit solution intents; use the index to freeze each intent's exact rendering bases, then read once only the deduplicated referenced sibling files. Project one complete `image_strategy` into each direction regardless of `recommend.image_usage`. Every candidate carries localized `name`, `rendering: custom`, `visual`, `mood`, and non-empty localized `behavior`. Mood includes a recognizable real-world analogy. All three must credibly serve their owning whole solution; rendering treatments and bases may coincide when other components express the direction difference. Do not force artificial safe / shifted / bold extremes. Image colors always inherit that direction's deck HEX roles; never add an image palette or alter deck colors to rescue a rendering.-Every direction is a `custom` rendering. It may use zero, one, or many index-selected bases: one may own a specialized treatment, while several must each own a distinct line, texture, depth, material, or mood contribution. Name every actual id in the visible behavior and read only those files after selection; reference count has no fixed cap, and a second basis is never required. If the direction is genuinely novel, name no basis and read none. Under a template it obeys inherited identity and application. Only a confirmed custom locks its edited behavior as `image_rendering_behavior`; when catalog material is actually used, also project the exact ids as `image_rendering_references`, otherwise omit that field. Unselected candidates remain recommendation-only. Do not write a separate fourth `custom_candidates.image_strategy`; ignore legacy `image_palette`.+Every direction is a `custom` rendering, unconstrained by how it relates to the catalog. It may use catalog material in any way or none, including carrying one complete preset treatment unchanged. Name every actual id in the visible behavior and read only those files after selection; when several are named, each owns a distinct line, texture, depth, material, or mood contribution. Reference count has no fixed cap, and a second basis is never required. With no catalog basis, name none and read none. Under a template it obeys inherited identity and application. Only a confirmed custom locks its edited behavior as `image_rendering_behavior`; when catalog material is actually used, also project the exact ids as `image_rendering_references`, otherwise omit that field. Unselected candidates remain recommendation-only. Do not write a separate fourth `custom_candidates.image_strategy`; ignore legacy `image_palette`.The UI hides these candidates while AI is not selected. If the user adds AI, it reveals the already-authored three without another backend recommendation; source selection never creates or rewrites rendering candidates. After confirmation, Image_Generator reads only the selected preset or exact custom references and must not blend unselected candidate identities.diff --git a/skills/ppt-master/references/strategist.md b/skills/ppt-master/references/strategist.mdindex d4f9901d..847bf830 100644--- a/skills/ppt-master/references/strategist.md+++ b/skills/ppt-master/references/strategist.md@@ -36,7 +36,7 @@ solution + production gate:| **1 — communication contract + template choice** | `primary_language` · `c` audience · open-ended communication intent · audience outcome · core message / delivery context (primary + optional secondary) / artifact afterlife · `content_divergence` (all prose fields may be blank) · `a` canvas · explicit `free_design` or `templates` choice and selected roots | confirmed together; candidate workspaces do not influence the communication recommendation || **2 — final solution + production** (authored once from the user's *actual* Stage 1) | reading mode (`delivery_purpose`, PPT only) · `d` mode + visual style · `b` page count · `e` color · `f` icon · `g` typography · `h` image source + generated-image rendering · conditional natural-language template application · conditional AI-image acquisition path · generation mode · refine-spec toggle · proactive speaker notes / custom animations / narration audio | derived as one coherent plan from the confirmed contract; internal template exporter modes remain hidden |-Do not force communication intent into one catalog label; Stage 1 records composite intent in prose. Editable prose fields are recommendation drafts, not required inputs: confirmation preserves current text and blanks; never repopulate a cleared field. Stage 2 confirms narrative spine, reading density, page budget, visual system, image direction, production mechanics, and how any installed template should be used. It never chooses or installs a template. Inspect only project-local template spec/prototypes, present one editable application plan, and keep exporter reuse/adherence internal. First author exactly three complete, project-fit solution directions from the confirmed contract and source; only then project each direction into mode, visual style, color, type, icons, and generated-image rendering for lower-level adjustment. Every direction projects a project-specific `custom` mode, `custom` visual style, and `custom` generated-image rendering; the fixed catalogs remain conservative lower-level single-select alternatives. All three must be viable and distinguishable as whole solutions, but do not force safe / shifted / bold archetypes, different catalog bases, or artificial extremes. After all three bundles are complete, compare them against the confirmed contract and source, choose the strongest overall fit, and write its actual zero-based index as `design_directions.selected` (`0`, `1`, or `2`); array order never determines preference. Every direction carries a complete generated-image rendering candidate even when AI imagery is not recommended; `recommend.image_usage` independently decides whether AI is proposed. Generated images inherit deck colors—there is no second image palette. Proactive defaults are speaker notes `true`, custom animations `false`, and narration audio `false`; a prior explicit user instruction overrides the matching recommendation, and effective narration audio requires effective speaker notes. Author each stage once; same-stage edits update only visible browser state through documented deterministic dependencies, without another AI/backend recommendation. Launch/derive/wait mechanics live in [`generate-pptx.md`](../workflows/generate-pptx.md) Step 4; item specs keep `a`–`h`.+Do not force communication intent into one catalog label; Stage 1 records composite intent in prose. Editable prose fields are recommendation drafts, not required inputs: confirmation preserves current text and blanks; never repopulate a cleared field. Stage 2 confirms narrative spine, reading density, page budget, visual system, image direction, production mechanics, and how any installed template should be used. It never chooses or installs a template. Inspect only project-local template spec/prototypes, present one editable application plan, and keep exporter reuse/adherence internal. First author exactly three complete, project-fit solution directions from the confirmed contract and source; only then project each direction into mode, visual style, color, type, icons, and generated-image rendering for lower-level adjustment. Every direction projects a project-specific `custom` mode, `custom` visual style, and `custom` generated-image rendering; the fixed catalogs remain conservative lower-level single-select alternatives. Each direction is one complete design authored top-down within the confirmed contract, never assembled bottom-up from catalog picks; three exist so a single recommendation cannot lock the user in, and the fixed catalogs stay the manual lower layer. Its custom projections are unrestricted by catalog relationship and may carry one preset unchanged. The three directions are plainly different designs at the whole-deck level before any field is written; that difference lives in the solutions, not in one designated field. Whichever components a direction's own design requires carry it — mode, visual style, rendering, color, type, or icons — any of them may, and none is required to differ. A different name, note, or reference count alone is not a difference, and projections identical on every component are not three solutions. Do not force safe / shifted / bold archetypes or artificial extremes. After all three bundles are complete, compare them against the confirmed contract and source, choose the strongest overall fit, and write its actual zero-based index as `design_directions.selected` (`0`, `1`, or `2`); array order never determines preference. Every direction carries a complete generated-image rendering candidate even when AI imagery is not recommended; `recommend.image_usage` independently decides whether AI is proposed. Generated images inherit deck colors—there is no second image palette. Proactive defaults are speaker notes `true`, custom animations `false`, and narration audio `false`; a prior explicit user instruction overrides the matching recommendation, and effective narration audio requires effective speaker notes. Author each stage once; same-stage edits update only visible browser state through documented deterministic dependencies, without another AI/backend recommendation. Launch/derive/wait mechanics live in [`generate-pptx.md`](../workflows/generate-pptx.md) Step 4; item specs keep `a`–`h`.**Default — continuity-aware whole solution (may override when a scene reset communicates better)**: Within active-profile invariants and before recommending page count or production mechanics, judge whether adjacent explanation beats can remain within one recognizable mental map while a visible state changes. Where that choice lowers cognitive switching and motion has a named communication job, let it shape the solution's narrative spine, page rhythm, visual approach, and enabled notes/narration segmentation, and recommend the existing `proactive_custom_animations: true`. This is one positive signal, not the only reason to enable animation; absent it, retain the existing default or other valid evidence. Topic or wording repetition alone is insufficient. A `Motion suggestion` remains optional advice and never changes the effective outcome.@@ -125,7 +125,7 @@ When authoring §IX, translate every purpose named in `communication_intent` intTwo independent layers, each locks one preset or `custom`. Output: `d. Mode: <mode> + Visual style: <visual_style>`.-> **Top-down custom direction construction.** Author three complete solution intents from the confirmed project contract and source before selecting any catalog basis; do not assemble three apparent solutions from independent mode/style/rendering picks. The fixed planning-capability batch may already be in context, but only the three mode/style/rendering indexes act as basis selectors. Use their summaries to freeze the exact reference ids for each direction, then read once only the deduplicated union of those referenced detail files and author the final behaviors. Every direction MUST serialize `mode: custom`, `visual_style: custom`, and `image_strategy.rendering: custom`, each with visible, non-empty behavior prose. A custom may use zero, one, or many catalog bases: one may specialize a strong dominant basis, several may divide distinct executable jobs, and a genuinely novel custom uses none. Reference count has no fixed cap and is an outcome, not a target; omit every basis whose contribution the behavior cannot state, and never add a second basis merely to make the result look synthesized. These three projections are the project-specific choices above the conservative fixed catalogs, not a fourth Custom proposal. Never glob a catalog, read an unselected sibling, or write bespoke prose as an enum value.+> **Top-down custom direction construction.** Author three complete solution intents from the confirmed project contract and source before selecting any catalog basis; do not assemble three apparent solutions from independent mode/style/rendering picks. The fixed planning-capability batch may already be in context, but only the three mode/style/rendering indexes act as basis selectors. Use their summaries to freeze the exact reference ids for each direction, then read once only the deduplicated union of those referenced detail files and author the final behaviors. Every direction MUST serialize `mode: custom`, `visual_style: custom`, and `image_strategy.rendering: custom`, each with visible, non-empty behavior prose. `custom` is not constrained by a catalog relationship: it may use catalog material in any way or none, including carrying one exact preset unchanged. References record only actual sources; one may carry the complete preset behavior, while several must each own a distinct executable contribution. Omit every basis whose contribution the behavior cannot state, and never add a decorative second basis. The three whole-deck directions MUST be plainly different designs before any field is written; complete each one's detail, then project that detail onto the fields. No component is the designated proof: whichever ones a direction's design requires carry the difference, any of them may, and mode, style, rendering, catalog bases, color, type, and icons are each free to coincide. A different name, note, or reference count alone is not a difference, and projections identical on every component are not three solutions. Where authoritative truth fixes some components, the remaining open ones carry it; when nothing is open, keep the projections identical and state that boundary instead of inventing a false difference. These three projections are the project-specific choices above the conservative fixed catalogs, not a fourth Custom proposal. Never glob a catalog, read an unselected sibling, or write bespoke prose as an enum value.#### Layer 1 — Communication mode@@ -136,8 +136,8 @@ The deck's **narrative + persuasion skeleton** — how the argument is organized**Source**:- User supplied their own outline / structure → preserve its facts and intended relationships, then apply the confirmed `content_divergence`. Treat an ordinary source outline as a Reference: regroup, reorder, or retitle when the communication contract benefits. Treat it as authoritative only when the user presents it as the final page plan or explicitly asks to preserve page order, titles, or wording; record that promoted boundary in `design_spec.md`. Still lock a mode for register, voice, and any permitted reshaping. `briefing` imposes the least if no particular "讲法" is intended.- Beautify / re-layout profile ([`beautify-pptx.md`](../workflows/profiles/beautify-pptx.md)) → the extracted source content is authoritative and **verbatim**, one step stricter than the user-outline case above. Each source slide becomes exactly one `§IX` page in source order; transcribe every content block word-for-word — never reshape / re-primary / condense / merge / split / reword. All three custom mode behaviors preserve that 1:1 boundary and may share `briefing` as their sole basis; do not manufacture narrative variation. Color (e) and typography (g) are whatever the user confirmed in the beautify plan — the source identity (theme or observed) by default, or a content / brand-aware alternative the beautify plan offered and the user picked — locked as truth. Charts / tables / images are regenerated from their extracted data in the inherited style: record only selected catalog references in §VII, keep unmatched chart/table plans in their §IX page blocks, and route pictures to §VIII. Data values stay frozen and the rendering is the deck's own; visuals are never carried over verbatim. Layout, hierarchy, rhythm, and visual rendering are what gets redesigned.-- Each direction crystallizes one project-specific cadence / posture in `mode_behavior`: it may specialize one preset, fuse several modes into a multi-act sequence, or be genuinely novel. A catalog-based direction retains only the exact ids it actually uses; a novel direction invents no basis. One deck locks one `custom` value, never several simultaneous modes.-- No user structure or cadence → derive each whole solution from the confirmed `communication_intent`, `audience_outcome`, source texture, and delivery context, then project its custom mode. The three directions may share a catalog basis when that is honestly best; distinguish them through project-specific behavior or other fields instead of forcing different bases.+- Each direction crystallizes one editable cadence / posture in `mode_behavior`. It is unconstrained by its relationship to the catalog and may use catalog material in any way or none, including carrying one preset unchanged. A catalog-based direction retains only the exact ids it actually uses; a direction using no catalog material invents no basis. One deck locks one `custom` value, never several simultaneous modes.+- No user structure or cadence → derive each whole solution from the confirmed `communication_intent`, `audience_outcome`, source texture, and delivery context, then project its custom mode. Directions may share catalog bases or the same mode behavior when their complete solutions remain plainly different; never force mode variation merely to separate the bundles.Record the confirmed mode and rationale in `design_spec.md` first, including every exact catalog basis when a selected custom uses any. Then project `- mode:` to `spec_lock.md`; for `custom`, also project `- mode_behavior:` and, only when catalog material is actually used, `- mode_references: <id>[, <id> ...]`. Executor reads only those exact references; an unreferenced novel custom follows the behavior directly.@@ -148,10 +148,10 @@ Record the confirmed mode and rationale in `design_spec.md` first, including eveThe deck's **visual aesthetic** — shape language, decoration density, whitespace rhythm, typographic character, texture. Anchors downstream fields e (Color), f (Icon), g (Typography), h (Image). Lock one preset from the catalog, or `custom`.**Source**:-- User named a style (chat / template / beautify) → it is truth: retain it as the required basis or inherited anchor in every custom behavior. Keep the three whole solutions meaningful by varying only fields the user left open; when visual variation is forbidden, the three style behaviors may be identical.-- No user description → author three project-fit whole solutions first, then project one complete custom aesthetic for each. Directions may share catalog bases when their overall systems still differ meaningfully. Do not force different bases, a safe-to-bold ladder, or one deliberately extreme option merely to manufacture variety. Give each direction a localized name and use its localized note as a compact, user-facing style summary. The note may reuse localized display labels from Confirm UI's `visual_styles` catalog (for example, `瑞士极简`, `柔和圆角`, or `编辑出版`) when they concisely describe the result, but these labels are optional vocabulary, not a selection constraint or required mapping. Use concise natural language wherever the catalog wording does not fit, and never force the nearest label. Keep the summary to one or two short sentences without exposing catalog ids or reference mechanics. The Confirm UI exposes these three project-specific styles above all 18 fixed manual alternatives.+- User named a style (chat / template / beautify) → it is truth: retain it as the required basis or inherited anchor in every custom behavior. Derive each direction's application through the style dimensions left open. When the user or template forbids all visual variation, the three style behaviors may be identical and the remaining open components carry the direction difference; state that boundary in the direction note instead of fabricating difference.+- No user description → author three project-fit whole solutions first, then project one complete custom aesthetic for each. A custom aesthetic is not constrained by its relationship to the catalog; it may use catalog material in any way or none, including carrying one fitting style unchanged. Directions may share catalog bases, and their `visual_style_behavior` values differ whenever the three designs genuinely differ in aesthetic rather than to satisfy a variation quota. Do not force different bases, a safe-to-bold ladder, or one deliberately extreme option merely to manufacture variety. Give each direction a localized name and use its localized note as a compact, user-facing style summary. The note may reuse localized display labels from Confirm UI's `visual_styles` catalog (for example, `瑞士极简`, `柔和圆角`, or `编辑出版`) when they concisely describe the result, but these labels are optional vocabulary, not a selection constraint or required mapping. Use concise natural language wherever the catalog wording does not fit, and never force the nearest label. Keep the summary to one or two short sentences without exposing catalog ids or reference mechanics. The Confirm UI exposes these three project-specific styles above all 18 fixed manual alternatives.-**Forbidden — a non-catalog name as `visual_style`**: every direction recommendation uses literal `custom` for `visual_style`; bespoke prose belongs only in `visual_style_behavior`, while optional `visual_style_references` contain only first-column catalog ids. A name from the `_index` "Paired rendering" column (`flat`, `vector-illustration`, `digital-dashboard`, `3d-isometric`, `corporate-photo`, …) is an image-rendering id, not a style reference. Generic words such as flat / modern / clean / simple / minimal are also insufficient behavior: use the index to choose an exact basis when applicable, then state the project-specific shape language, composition, density, whitespace, typography, and texture.+**Forbidden — a non-catalog name as `visual_style`**: every direction recommendation uses literal `custom` for `visual_style`; bespoke prose belongs only in `visual_style_behavior`, while optional `visual_style_references` contain only first-column catalog ids. A name from the `_index` "Paired rendering" column (`flat`, `vector-illustration`, `digital-dashboard`, `3d-isometric`, `corporate-photo`, …) is an image-rendering id, not a style reference. Generic words such as flat / modern / clean / simple / minimal are also insufficient behavior: use the index to choose an exact basis when applicable, then state the executable shape language, composition, density, whitespace, typography, and texture carried into this direction. Those rules may match one preset exactly; do not invent a difference merely to justify `custom`.**Carries no color.** A visual style governs how the deck's HEX (locked at `e`) is *used* — never which colors, same discipline as [`image-renderings`](./image-renderings/_index.md). When the deck has AI images, prefer the style's paired rendering so layout and illustration share one aesthetic.@@ -598,7 +598,7 @@ includes transitions.**Contextual extension**: derived paint or sparse local font/color garnish may stay in one SVG while non-structural and non-recurring. New base/semantic colors, structural/recurring fonts, resources, or recurring cross-page identity patterns require upstream repair; a page-local §VIII preferred image pattern follows [`executor-image.md`](./executor-image.md) and may change during realization. Executor never reverse-projects a local choice as planning fact. Promote recurring garnish upstream before reuse, read back and validate the affected planning fragments, and never add values to silence a comparison.- **Communication trace is mandatory**: Keep the full confirmed communication contract in `design_spec.md §I`, then project only `audience`, `objective`, `core_message`, and canonical `consumption_mode` into `spec_lock.md communication`. Write `objective` as one concise execution sentence that preserves both the confirmed `communication_intent` and the success condition in `audience_outcome`; do not copy `delivery_context`, `artifact_afterlife`, dates, provenance, or conflict-resolution commentary into the lock. Before finalizing §IX, check that every named purpose has at least one outline obligation and **every Slide block**, including cover / divider / closing pages, has an `Audience move` that advances the global outcome. A page that advances no purpose or outcome should be merged, rewritten, or cut. `project_manager.py validate` and `svg_quality_checker.py` enforce the compact lock fields and per-page move presence, not their subjective quality.- - **Custom behavior is concise and executable**: For confirmed `custom` mode or visual style, project one resolved `mode_behavior` / `visual_style_behavior` sentence or short paragraph. When the direction actually combines or borrows catalog entries, also project the exact, comma-separated `mode_references` / `visual_style_references`; omit the field for a genuinely novel direction and never fabricate a nearby reference. Preserve the confirmed direction, reference locked role names such as `colors.primary` when needed, and omit selection history, contradictions, precedence explanations, or other Design Spec provenance. Executor reads these fields from the retained lock and loads every referenced catalog entry once per valid context.+ - **Custom behavior is concise and executable**: For confirmed `custom` mode or visual style, project one resolved `mode_behavior` / `visual_style_behavior` sentence or short paragraph. When the direction actually uses catalog entries, also project the exact, comma-separated `mode_references` / `visual_style_references`; omit the field when it uses none and never fabricate a nearby reference. Preserve the confirmed direction, reference locked role names such as `colors.primary` when needed, and omit selection history, contradictions, precedence explanations, or other Design Spec provenance. Executor reads these fields from the retained lock and loads every referenced catalog entry once per valid context.- **page_rhythm is mandatory**: Based on the page list in §IX Content Outline, assign each page one of `anchor` / `dense` / `breathing`. This is what breaks the uniform "every page is a card grid" feel. New locks may not omit the section; consumer omission behavior is owned by [`executor-base.md`](executor-base.md) §2.1.- **Fact IDs and scenario labels are mandatory when applicable**: Read any `sources/*.facts.json`. For each §IX page, list the stable IDs actually used; never cite an ID whose claim is absent from the page. Mark invented KPIs/targets/internal ratios as `Data class: scenario` and state which values are scenario data. Executor carries external sources into notes/footnotes and renders a visible scenario label for scenario figures.- **Mandatory — whole-roster rhythm check**: During the same §IX composition, compare neighbors and section arcs to judge whether chapter entries visibly reset, extended same-density runs are intentional, extended same-carrier or same-topology runs form an intentional semantic sub-arc, repeated dominant geometry carries a continuity job, any qualifying §6.1 visible-state sequence preserves a recognizable mental map while making its next semantic change legible, each section follows a mode-fitting progression—including framework → explanation/evidence → judgment/action when it serves the objective—and the final arc resolves the communication objective before a genuine ending lowers information load. Same section, equal weight or density, one style, and prior-page precedent do not establish a semantic sub-arc. Repair the existing roster, `Layout` recommendations, and `page_rhythm` choices in place. This is judgment, not quota; preserve intentional continuity, legitimately all-`dense` material, and 1:1/literal order. Do not invent filler pages to manufacture rhythm; a `breathing` page marks a meaningful pause—chapter transition, standalone emphasis, or SCQA bridge—and must stand alone. Create no field, lock row, artifact, or second review/execution pass.diff --git a/skills/ppt-master/references/visual-styles/_index.md b/skills/ppt-master/references/visual-styles/_index.mdindex fd37e556..e1f234bc 100644--- a/skills/ppt-master/references/visual-styles/_index.md+++ b/skills/ppt-master/references/visual-styles/_index.md@@ -105,13 +105,13 @@ an exact authoring preset here.----## 3. Escape hatch — `custom`+## 3. Editable `custom` projection-Each coordinated Default Stage-2 direction authors one visible, non-empty `custom` aesthetic whose paragraph names project-specific shape language, composition geometry, decoration density, whitespace, typographic character, and texture — **no HEX, no color names as values**. Use this index to freeze any catalog bases before reading their detail files. A template-backed synthesis stays inside the inherited identity and confirmed application plan. Record the confirmed current aesthetic in the Design Spec first, then project `- visual_style: custom` plus `- visual_style_behavior:`. The 18 fixed styles remain lower-level single-select alternatives; do not create a fourth AI-custom proposal.+Each coordinated Default Stage-2 direction authors one visible, non-empty `custom` aesthetic whose paragraph names its executable shape language, composition geometry, decoration density, whitespace, typographic character, and texture — **no HEX, no color names as values**. `custom` is not constrained by its relationship to the catalog: it may use catalog material in any way or none, including carrying one fitting preset unchanged. The three complete directions are plainly different designs; this axis expresses that difference whenever their aesthetics genuinely differ, and catalog bases may coincide. A different name, note, reference count, or palette alone is not another aesthetic. Style behaviors may also coincide when other components carry the direction difference or authoritative user/template truth forbids visual variation; state that boundary in the direction note instead of fabricating difference. Use this index to freeze any catalog bases before reading their detail files. A template-backed direction stays inside the inherited identity and confirmed application plan. Record the confirmed current aesthetic in the Design Spec first, then project `- visual_style: custom` plus `- visual_style_behavior:`. The 18 fixed styles remain lower-level single-select alternatives; do not create a fourth AI-custom proposal.Quick does not display a candidate spectrum. It reads this index, resolves one preset or custom behavior, then reads only the selected detail files and persists nothing.-**Mandatory — select before detail reading**: Freeze every catalog source actually used from this index, then read only those exact files before writing the behavior. A custom may use zero, one, or many sources: keep one when it owns the whole specialized aesthetic, or include every style that contributes a distinct executable job across shape language, composition, decoration, whitespace, typography, or texture. Reference count has no fixed cap; count is an outcome, not a target. A coherent three-basis direction may assign `swiss-minimal` to grid and whitespace, `soft-rounded` to selective surface contours and elevation, and `editorial` to evidence hierarchy and rules. Default persists every actual id as `visual_style_references`; Quick retains them only in active context. Omit every source whose contribution cannot be stated, never add a second merely to imply synthesis, and do not open candidates for comparison after this gate. A genuinely new aesthetic names and reads no catalog source.+**Mandatory — select before detail reading**: Freeze every catalog source actually used from this index, then read only those exact files before writing the behavior. One source may supply the complete aesthetic unchanged; when several are named, each contributes a distinct executable job across shape language, composition, decoration, whitespace, typography, or texture. Reference count has no fixed cap; count is an outcome, not a target. A coherent three-basis direction may assign `swiss-minimal` to grid and whitespace, `soft-rounded` to selective surface contours and elevation, and `editorial` to evidence hierarchy and rules. Default persists every actual id as `visual_style_references`; Quick retains them only in active context. Omit every source whose contribution cannot be stated, never add a second merely to imply synthesis, and do not open candidates for comparison after this gate. A custom using no catalog source names and reads none.---diff --git a/skills/ppt-master/scripts/docs/confirm_ui.md b/skills/ppt-master/scripts/docs/confirm_ui.mdindex d7cd9c1a..9367f955 100644--- a/skills/ppt-master/scripts/docs/confirm_ui.md+++ b/skills/ppt-master/scripts/docs/confirm_ui.md@@ -313,11 +313,11 @@ template-selection receipt.- **Closed enumerable** — PPT reading mode (`delivery_purpose` compatibility key), generation mode / refine spec, plus AI source only when image usage includes `ai`. These have no Custom box; out-of-catalog values snap back to the recommended option.- **Proactive execution booleans** — Final Stage 2 carries top-level `proactive_speaker_notes`, `proactive_custom_animations`, and `proactive_narration_audio` values. Defaults are `true`, `false`, and `false`, respectively. They control what the Agent does proactively only when the user has not explicitly instructed otherwise; the latest explicit user instruction always wins. These three values are raw confirmation evidence: the UI and server neither couple nor rewrite them, and every boolean combination is valid. When narration audio is enabled, Strategist later resolves the effective Speaker Notes outcome to enabled and records `Narration Audio dependency` as its Design Spec provenance. Disabling proactive custom animation does not suppress the Strategist's advisory motion recommendations.- **Open prose** — `audience`, `communication_intent`, `audience_outcome`, `core_message`, `delivery_context`, `artifact_afterlife`, `content_divergence`, and `page_count`. `communication_intent` may carry several purposes plus priority / sequence; common paths appear only as help text. `delivery_context` states one primary presenter-led / reader-led / hybrid / recorded-self-running context plus optional secondary use; a hybrid recommendation names which context leads. `content_divergence` is the source-treatment axis. `page_count` may be a range here; Strategist resolves the exact §IX roster, leaving Executor no pagination latitude.-- **Coordinated generative directions** — `design_directions` carries exactly three complete candidates authored top-down from the project contract. Each has a unique stable id and bundles `custom` mode, `custom` visual style, color, typography, icon id, and `custom` generated-image rendering regardless of recommended image source. Its localized note is a compact, user-facing style summary. It may reuse localized display labels from `catalogs.visual_styles` when they describe the result concisely, but those labels are optional vocabulary rather than a selection constraint or required mapping. Otherwise it uses concise natural language and never forces the nearest label. The summary stays within one or two short sentences and does not expose catalog ids or reference mechanics. All three must be viable as whole solutions; they do not need different catalog bases or forced safe / shifted / bold archetypes. After completing all three bundles, Strategist compares them against the confirmed contract and source, then writes the strongest overall fit's zero-based index to `selected`; array position does not determine preference. That bundle becomes the initial default and applies its three custom projections coherently. The page can still render legacy top-level `color`, `typography`, and `image_strategy` candidates, but new staged recommendations use the coordinated bundle.+- **Coordinated generative directions** — `design_directions` carries exactly three complete candidates authored top-down from the project contract. Each has a unique stable id and bundles `custom` mode, `custom` visual style, color, typography, icon id, and `custom` generated-image rendering regardless of recommended image source. Its localized note is a compact, user-facing style summary. It may reuse localized display labels from `catalogs.visual_styles` when they describe the result concisely, but those labels are optional vocabulary rather than a selection constraint or required mapping. Otherwise it uses concise natural language and never forces the nearest label. The summary stays within one or two short sentences and does not expose catalog ids or reference mechanics. Each candidate is one complete design authored top-down within the confirmed contract, never assembled bottom-up from catalog picks; three exist so a single recommendation cannot lock the user in, while the fixed catalogs stay the manual lower layer. Its custom projections are unrestricted by catalog relationship and may carry one preset unchanged. The three candidates are plainly different designs at the whole-deck level, and that difference lives in the solutions rather than in one designated field: whichever components a candidate's design requires carry it, and mode, visual style, rendering, catalog bases, color, typography, and icons are each free to coincide. A different name, note, or reference count alone is not a difference, and candidates identical on every component are not three solutions. Do not force safe / shifted / bold archetypes. Where authoritative user/template truth fixes some components, the remaining open ones carry the difference; identical projections are valid only when nothing is open, stated as that boundary in the direction note. After completing all three bundles, Strategist compares them against the confirmed contract and source, then writes the strongest overall fit's zero-based index to `selected`; array position does not determine preference. That bundle becomes the initial default and applies its three custom projections coherently. The page can still render legacy top-level `color`, `typography`, and `image_strategy` candidates, but new staged recommendations use the coordinated bundle.Direction-local custom projections apply to mode, visual style, and generated-image rendering; all three are editable after selection and a selected custom value cannot be blank. The original recommendation remains immutable so the active whole-direction card can explicitly restore every edited component without making an ordinary card click destructive. Legacy standalone `custom_candidates` remain readable but are optional and are not authored in new files. Color / typography keep their existing manual Custom cards. Image usage uses source ids plus `image_notes`; closed sets have no Custom path.-**Stage-2 catalog read gate.** Before choosing component bases, Strategist reads only `modes/_index.md`, `visual-styles/_index.md`, and `image-renderings/_index.md`. It authors the three whole solution intents first, freezes every basis id from those indexes, and only then reads the deduplicated selected detail files before completing the custom behaviors. Unselected sibling files never enter context; a novel custom reads none.+**Stage-2 catalog read gate.** Before choosing component bases, Strategist reads only `modes/_index.md`, `visual-styles/_index.md`, and `image-renderings/_index.md`. It authors the three whole solution intents first, freezes every basis id from those indexes, and only then reads the deduplicated selected detail files before completing the custom behaviors. `custom` has no required catalog relationship: one exact preset may supply an entire editable projection unchanged, several may contribute distinct jobs, or no preset may be used. These are examples, not a closed classification. The behavior remains non-empty; omit every source whose contribution it cannot state and never add a decorative second basis. Unselected sibling files never enter context.**Stage-1 current-value contract.** Each editable prose box starts with the Strategist's recommendation, if one exists. The user may retain, revise, or clear it; no Stage-1 prose field has a non-empty validation gate. On confirmation, the browser submits the current strings and the server preserves them through every later stage and the final `result.json`, including `""`. Blank means no explicit user constraint and may cause downstream default judgment, but it never causes the initial recommendation to be restored. A profile-declared `locked: true` field is read-only and remains the sole exception.@@ -583,9 +583,9 @@ custom` + `visual_style_behavior`, or `image_strategy.rendering: custom` +`behavior`. During Design Spec and lock authoring, Strategist projects optional`mode_references`, `visual_style_references`, or`image_rendering_references` only when that confirmed behavior actually uses-named catalog sources. These lists have no fixed item limit; each item must own-a distinct executable contribution, while genuinely novel custom behavior has-no reference list.+named catalog sources. These lists have no fixed item limit. One item may carry+the complete preset behavior unchanged; with several, each owns a distinct+executable contribution. Genuinely novel custom behavior has no reference list.The Stage-1 intermediate write retains the communication contract for Stage 2.**Final-result consumption contract.** A final result is the user-confirmed input contract for the Strategist's Design Spec, not another recommendation input. After the final wait, Generate Step 4 reads the complete final object exactly once and retains it while Strategist writes and audits `design_spec.md` against every explicitly present field. Normal lock authoring and downstream execution do not reopen `result.json`; the completed Design Spec is the durable authority. Only after that audit passes does Strategist author `spec_lock.md` from the Design Spec plus current execution context, selecting stable anchors and routing rather than copying every field or enumerating every legal color/font. Every value must be consumed at the semantic type owned by [`strategist.md`](../../references/strategist.md) §1 and its field owner: do not omit or substitute it, and do not silently strengthen or weaken its type. If a confirmed requirement cannot be honored, the owning workflow reports or pauses under failure recovery; it never deletes the requirement to keep the pipeline moving.diff --git a/skills/ppt-master/workflows/generate-pptx.md b/skills/ppt-master/workflows/generate-pptx.mdindex 4738c0a2..80936854 100644--- a/skills/ppt-master/workflows/generate-pptx.md+++ b/skills/ppt-master/workflows/generate-pptx.md@@ -624,7 +624,7 @@ Read ${SKILL_DIR}/references/visual-styles/<resolved-id>.md # one preset id, orKeep the core's shared visual-quality defaults and `svg-effects.md` §6.1 Visual Job Router active during page authoring; they are not passive compatibility reading. Explicit user/template requirements and the locked style override compatible aesthetic defaults, never technical Required / Forbidden boundaries.-> Read only the role core, always-on construction references, exact locked detail files, and conditionally triggered modules below. The selection indexes remain planning-only. A preset reads its one locked file. For `custom`, read only the exact bases named by optional `mode_references` / `visual_style_references`, then synthesize them under the corresponding behavior. If absent, treat the direction as genuinely novel and read no preset file. Do not infer adjacent bases, glob a catalog, or blend unselected identities.+> Read only the role core, always-on construction references, exact locked detail files, and conditionally triggered modules below. The selection indexes remain planning-only. A preset reads its one locked file. For `custom`, read only the exact bases named by optional `mode_references` / `visual_style_references`: apply one under the corresponding behavior, or synthesize several by their stated contributions. If absent, read no preset file and follow the behavior directly. Do not infer adjacent bases, glob a catalog, or blend unselected identities.| Deterministic trigger | Additional references ||---|---|diff --git a/skills/ppt-master/workflows/profiles/quick-generate.md b/skills/ppt-master/workflows/profiles/quick-generate.mdindex d70fe807..0e57ae3a 100644--- a/skills/ppt-master/workflows/profiles/quick-generate.md+++ b/skills/ppt-master/workflows/profiles/quick-generate.md@@ -513,9 +513,9 @@ not route among them one file at a time:[`semantic-svg.md`](../../references/semantic-svg.md),and [`executor-structure.md`](../../references/executor-structure.md). Retainonly the mode/style detail files selected during one-pass design resolution and-realize that chosen direction. Exact `*_references` define the complete basis-of a custom synthesis; a genuinely novel custom follows its resolved behavior-without reading or inventing a nearby preset.+realize that chosen direction. Exact `*_references` define the catalog material+actually used by a custom: apply one basis under its behavior, synthesize several+by their stated contributions, or follow the behavior directly when none exist.Do not load `executor-base.md`: it owns Default's persisted-plan handoff,first-page gate, and completion routing. Excluding that file is not a capabilitydiff --git a/skills/ppt-master/workflows/stages/resume-execute.md b/skills/ppt-master/workflows/stages/resume-execute.mdindex c9427a59..f996ece7 100644--- a/skills/ppt-master/workflows/stages/resume-execute.md+++ b/skills/ppt-master/workflows/stages/resume-execute.md@@ -98,7 +98,7 @@ Then continue the documented pipeline:- When §X records a final/literal narration script, read the verified frozen `notes/total.md` once and retain its page segments through SVG authoring and the late notes validation- If resuming mid-deck, read the latest completed SVG and current image metadata after their required paths have been verified-- Read the remaining Step 6 construction core exactly as listed in [`generate-pptx.md`](../generate-pptx.md), then read one locked preset file or only the exact `*_references` of a custom synthesis; never reopen or glob the mode or visual-style catalogs, and load only the branches selected by the condition table+- Read the remaining Step 6 construction core exactly as listed in [`generate-pptx.md`](../generate-pptx.md), then read one locked preset file or only the exact `*_references` of a custom; apply one basis under its behavior or synthesize several by their stated contributions. Never reopen or glob the mode or visual-style catalogs, and load only the branches selected by the condition table- For each page, make the mandatory Structure decision from retained §IX after its content/communication move is established and before any geometry; a `yes` result loads `executor-structure.md` before realization and creates no artifact or lock row- Design Parameter Confirmation- When structured, read the template Design Spec and each selected prototype once; retain unchanged references in the fresh context. A later bounded repair follows [`executor-base.md`](../../references/executor-base.md) §2.1 only while that context remains valid and uncompacted