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
refactor(vocab): unify confirmation stage schema and pipeline session naming
hugohe30d78c8afa1dd
diff --git a/AGENTS.md b/AGENTS.mdindex 2f5513fd..286908ca 100644--- a/AGENTS.md+++ b/AGENTS.md@@ -8,7 +8,7 @@ This file is the project entry point for general AI agents.PPT Master is an AI-driven presentation generation system. Multi-role collaboration (Strategist → Image_Generator → Executor) converts source documents (PDF/DOCX/URL/Markdown) into natively editable PPTX with real PowerPoint shapes (DrawingML).-**Core Pipeline**: `Source Document → Create Project → [Template] → Strategist Eight Confirmations → [Image_Generator] → Executor Live Preview → Quality Check → Post-processing → Export PPTX`+**Core Pipeline**: `Source Document → Create Project → [Template] → Strategist confirmation stage → [Image_Generator] → Executor Live Preview → Quality Check → Post-processing → Export PPTX`**Route selection authority**: [`skills/ppt-master/workflows/routing.md`](skills/ppt-master/workflows/routing.md) owns the complete route matrix. The hard boundaries below stay inline because they bypass or redirect the main pipeline and are the most expensive to misroute.@@ -53,7 +53,7 @@ python3 skills/ppt-master/scripts/project_manager.py validate <project_path># Icon selection — copy chosen library icons into <project>/icons/ (missing names reported + non-zero = re-pick)python3 skills/ppt-master/scripts/icon_sync.py <project_path> <lib/name> [<lib/name>...]-# Step 4 Eight Confirmations — interactive visual page (default auto-launch; chat fallback)+# Step 4 Strategist confirmation stage — interactive visual page (default auto-launch; chat fallback)python3 skills/ppt-master/scripts/confirm_ui/server.py <project_path> --daemon --wait# Image tools and SVG quality checkdiff --git a/docs/faq.md b/docs/faq.mdindex 4b8324f2..5b9ed638 100644--- a/docs/faq.md+++ b/docs/faq.md@@ -162,9 +162,9 @@ If generation feels slow, check your model's token throughput. The bottleneck isDefault recommendation: **continuous one-shot generation**. 10–15 page decks fit comfortably in a 200K window, and cross-page visual consistency is best when the Executor can see prior pages in the same session (it actively aligns style, font sizes, and rhythm).-Only when signals are heavy (≥ 18 pages, thick source material, or `topic-research` ran with substantial web-fetch accumulation) does the AI surface an optional **two-stage (split mode)** hint at the Strategist phase: Phase A (eight confirmations + image acquisition) ends in the current chat; you open a fresh chat window and type `继续生成 projects/<project_name>` (or "resume execution projects/<project_name>") to enter Phase B (SVG generation + export). The new session reloads `design_spec` / `spec_lock` / `sources` / `images` from disk and continues from there.+Only when signals are heavy (≥ 18 pages, thick source material, or `topic-research` ran with substantial web-fetch accumulation) does the AI surface an optional **split mode** hint at the Strategist phase: the planning session (Strategist confirmation stage + image acquisition) ends in the current chat; you open a fresh chat window and type `继续生成 projects/<project_name>` (or "resume execution projects/<project_name>") to enter the execution session (SVG generation + export). The new session reloads `design_spec` / `spec_lock` / `sources` / `images` from disk and continues from there.-Split mode is a **compromise** — it pays ~6K tokens (re-reading SKILL.md) to drop 60–200K of Phase A noise, then reuses the freed budget in Phase B to re-read `sources/` for richer slide content. **Not needed when signals are normal**; the hint won't appear, and you can always ignore it and stay in continuous mode.+Split mode is a **compromise** — it pays ~6K tokens (re-reading SKILL.md) to drop 60–200K of planning-session noise, then reuses the freed budget in the execution session to re-read `sources/` for richer slide content. **Not needed when signals are normal**; the hint won't appear, and you can always ignore it and stay in continuous mode.## Q: Can I preview or fix individual pages before the full export?diff --git a/docs/roadmap.md b/docs/roadmap.mdindex 9f80d55b..d810b000 100644--- a/docs/roadmap.md+++ b/docs/roadmap.md@@ -52,8 +52,8 @@ The past two months' structural capability growth. Single flags / incremental po- **visual-style catalog grows to 18, aligned with image-renderings + examples reclaimed** — first four distilled from the [examples library](../examples/) (`brutalist` / `blueprint` / `memphis` / `zine`), then six filling in the hand-drawn / textured renderings that have a layout twin (`sketch-notes` / `ink-notes` / `chalkboard` / `paper-cut` / `vintage-poster` / `pixel-art`), then four more reclaimed from still-uncovered example aesthetics: `ink-wash` (rice-paper whitespace, from 藏拙 / 李子柒) · `glassmorphism` (dark frosted-glass + gradient light, from glassmorphism_demo, split out of soft-rounded) · `photo-editorial` (full-bleed photography dominates, text captions, from Pritzker / fashion_weekly) · `data-journalism` (Bloomberg/Economist multi-column micro-charts + sidebars, from global_ai_capital). The catalog is regrouped into five families (corporate-product / editorial-publication / expressive-print / hand-drawn-brush / specialty). **The test**: a rendering earns a visual-style twin only when it defines a whole-page layout language, not merely how an inserted image looks — so photo-*led composition* gets `photo-editorial` (paired with corporate-photo), while purely atmospheric renderings (nature / warm-scene / fantasy-animation) stay imagery-only and just pair with a layout style. Zero HEX, Reference strength throughout- **mode catalog grows to 5: add `briefing`** — fills the "neutral information delivery" cell: no thesis, no story, no teaching, no spectacle — topic titles, even weight, complete and scannable, for status updates / reference decks / catalogs / meeting packs / FAQs that inform without arguing. The five now partition presentation *intent* more nearly MECE: persuade (pyramid) · tell a story (narrative) · teach (instructional) · impress (showcase) · simply inform (briefing). `_index` gains a `briefing` vs `pyramid` tie-breaker ("if you'd have to invent a thesis to fit pyramid, it's briefing"). Five presets plus a `custom` escape hatch for a bespoke direction none of them captures (a special cadence, a multi-mode fusion, a particular posture) — either user-requested or Strategist-recommended, confirmed like every lock; a deck always locks one value, and a fusion is one `custom` describing the acts. The only thing to avoid is defaulting to `custom` when a preset genuinely fits. Mirrors the truth-precedence rule that a user-authored outline or direction overrides the mode- **mode / visual-style system validated on real decks + four calibration tightenings landed** — the 5 modes + 18 visual-styles + the `custom` escape hatch were exercised on five covering decks (briefing×data-journalism / narrative×photo-editorial / instructional×chalkboard / showcase×glassmorphism / custom×zine, with the narrative deck going through the AI-image branch): **zero selection misses** (all four Close-calls tie-breakers fired and held against the real grey-zone pulls), **discipline fully honored** (zero HEX / Reference strength / whole-page layout language), **the `custom` mechanism works** (a `mode_behavior` prose paragraph carried a 10-page generation and reads as plain language at confirmation, not a bare token), **mode ⟂ visual_style holds** (any combination without bleed, including a positive check of the "keynote/launch = mode not style" route), export 5/5 decks × every page with 0 failures. Tightened four spots from the real signal: `strategist §e` now locks the full neutral tier set the visual_style implies up front (killing the mid-deck color top-ups seen in three consecutive decks), `executor-base §1` re-skins chart/layout template gradients-shadows-fills to the deck's visual_style (templates supply structure, not skin; mirror templates stay verbatim per §1.1), `briefing §1` makes `core_message` state coverage rather than a claim (briefing-only; global §IX assertion semantics stay for narrative/instructional/pyramid), and `svg_quality_checker` fixes font-family drift false positives (delimiter-matched capture + stack normalization) and drops the font-size ceiling for showcase mode and poster visual-styles-- **Optional spec-review checkpoint shipped** ([`workflows/refine-spec.md`](../skills/ppt-master/workflows/refine-spec.md)) — an opt-in pause after the Eight Confirmations: when the user explicitly asks (default OFF), the Strategist produces the full `design_spec.md` + `spec_lock.md`, then stops so the user can revise any part of the spec (outline / color / typography / layout / image strategy / page rhythm) before generation, keeping both files in sync. Same shape as the split-mode note — it never fires on its own, the default pipeline is unchanged, and it surfaces as one opt-in line inside the Eight Confirmations. Review lenses (logical clarity / information density / focus / register / emotional resonance / chapter balance + the design dimensions) give a direction only, never a numeric threshold (`Reference` strength). Prompted by [@cuberoocp](https://github.com/cuberoocp) in [issue #173](https://github.com/hugohe3/ppt-master/issues/173)-- **Interactive visual Eight Confirmations page (Step 4)** ([`scripts/confirm_ui/server.py`](../skills/ppt-master/scripts/confirm_ui/server.py), field schema [`scripts/docs/confirm_ui.md`](../skills/ppt-master/scripts/docs/confirm_ui.md)) — the Eight Confirmations move from chat-only to a browser page that auto-launches by default at Step 4: enumerable fields (canvas / mode / visual_style / icons / image usage / formula & generation policy) list common options from `catalogs.json`, while generative fields present candidates — color swatches, live font previews (CJK / Latin previewed independently), and AI-image rendering × palette picks; plus live swatch feedback on custom HEX, a live combined color × typography preview, and a body-size range hinted from the chosen canvas. It **shares port `5050`** with the Step 6 Live Preview (they never run at once — the page auto-`--shutdown`s at the end of Step 4 to free the port; if the port is busy the launcher auto-advances to the next free one). On principle **chat stays the canonical channel and the page is a convenience layer**: the page's `result.json` is authoritative over the recommendations and is consumed in place downstream (image plan / `image_strategy` / fonts / the split and refine-spec toggles all read from it), and any failure to open / timeout / headless host degrades losslessly to the chat-summary fallback+- **Optional spec-review checkpoint shipped** ([`workflows/refine-spec.md`](../skills/ppt-master/workflows/refine-spec.md)) — an opt-in pause after the Strategist confirmation stage: when the user explicitly asks (default OFF), the Strategist produces the full `design_spec.md` + `spec_lock.md`, then stops so the user can revise any part of the spec (outline / color / typography / layout / image strategy / page rhythm) before generation, keeping both files in sync. Same shape as the split-mode note — it never fires on its own, the default pipeline is unchanged, and it surfaces as one opt-in line inside the Strategist confirmation stage. Review lenses (logical clarity / information density / focus / register / emotional resonance / chapter balance + the design dimensions) give a direction only, never a numeric threshold (`Reference` strength). Prompted by [@cuberoocp](https://github.com/cuberoocp) in [issue #173](https://github.com/hugohe3/ppt-master/issues/173)+- **Interactive visual Strategist confirmation stage page (Step 4)** ([`scripts/confirm_ui/server.py`](../skills/ppt-master/scripts/confirm_ui/server.py), field schema [`scripts/docs/confirm_ui.md`](../skills/ppt-master/scripts/docs/confirm_ui.md)) — the Strategist confirmation stage moves from chat-only to a browser page that auto-launches by default at Step 4: enumerable fields (canvas / mode / visual_style / icons / image usage / formula & generation policy) list common options from `catalogs.json`, while generative fields present candidates — color swatches, live font previews (CJK / Latin previewed independently), and AI-image rendering × palette picks; plus live swatch feedback on custom HEX, a live combined color × typography preview, and a body-size range hinted from the chosen canvas. It **shares port `5050`** with the Step 6 Live Preview (they never run at once — the page auto-`--shutdown`s at the end of Step 4 to free the port; if the port is busy the launcher auto-advances to the next free one). On principle **chat stays the canonical channel and the page is a convenience layer**: the page's `result.json` is authoritative over the recommendations and is consumed in place downstream (image plan / `image_strategy` / fonts / the split and refine-spec toggles all read from it), and any failure to open / timeout / headless host degrades losslessly to the chat-summary fallback- **A batch of source-conversion fidelity gains** — less information dropped as source material enters the pipeline: `doc_to_md` converts Word OMML / Office Math equations to inline LaTeX, `pdf_to_md` recognizes `Figure N |` pipe-delimited captions, and `ppt_to_md` preserves hyperlinks already in the source deck (run-level external `[text](url)` / slide-internal jumps `[text](#slide-N)` / shape-level clicks, with dangerous-scheme filtering and anchor-text Markdown escaping) and transcribes native chart data into Markdown tables (values survive the conversion instead of collapsing to a single picture). Caption recognition is based on [@suay1113](https://github.com/suay1113)'s [PR #191](https://github.com/hugohe3/ppt-master/pull/191); hyperlink preservation is distilled from [@ZhaoZuohong](https://github.com/ZhaoZuohong)'s [PR #155](https://github.com/hugohe3/ppt-master/pull/155)- **Content-faithful PPT beautification / re-layout shipped** ([`workflows/beautify-pptx.md`](../skills/ppt-master/workflows/beautify-pptx.md)) — mirror of `template-fill`: template-fill reuses a deck's design with new content, beautify keeps the content and redoes the layout. Given an existing PPTX, **every text string is preserved verbatim (nothing added / removed / reworded)**; it **extracts and inherits the source deck's visual identity (palette / fonts, `theme` or `observed` offered as two confirm-page candidates)** and redoes only layout / hierarchy / whitespace, strict 1:1 page count and order, with charts / tables regenerated natively from extracted data (values frozen) and source pictures re-laid-out. Technically still "generate a native, editable PPTX from scratch" (`ppt_to_md` extracts content → main pipeline → a brand-new deck), not a patch over the original, so it stays clear of Non-goals #53. Adds `beautify_identity.py` / `beautify_inventory.py`; the confirm page is seeded from the source for the user to review. Honest v1 ceiling: it does not relieve information overload (a crowded page improves within itself; true re-pagination is the main pipeline), does not guarantee coordinate-level paste-back, and combo / dual-axis / waterfall charts lose the un-captured plotsdiff --git a/docs/technical-design.md b/docs/technical-design.mdindex 1f5dc5b3..c24f0b31 100644--- a/docs/technical-design.md+++ b/docs/technical-design.md@@ -27,7 +27,7 @@ User Input (PDF/DOCX/XLSX/PPTX/URL/Markdown/topic text)Trigger only on an explicit template directory path containing design_spec.md kind: brand/layout/deckRaw PPTX template requests route to template-fill; reusable SVG templates are created by create-template first↓-[Strategist] - two-tier Eight Confirmations & Design Specifications → design_spec.md + spec_lock.md+[Strategist] - three-stage Strategist confirmation stage & Design Specifications → design_spec.md + spec_lock.md↓[Image Acquisition] (when any resource row needs AI generation, web search, or slicing)↓@@ -246,7 +246,7 @@ The three template kinds own different segments of the design contract:| Kind | Owns | Typical contents | Effect on Strategist ||---|---|---|---|| `brand` | identity | colors, typography, logo, voice, icon style | locks identity; structure remains free |-| `layout` | structure | canvas, page structure, page types, SVG roster | locks structure; identity is confirmed in the Eight Confirmations |+| `layout` | structure | canvas, page structure, page types, SVG roster | locks structure; identity is confirmed in the Strategist confirmation stage || `deck` | identity + structure + template overview | full replica-style package | locks the full template grammar, with only content-specific choices left |When several paths are supplied, fusion is segment-level, not field-level. A brand overrides the identity segment, a layout overrides the structure segment, and a deck supplies the middle/template-overview segment. Same-kind conflicts are surfaced as conflicts rather than resolved by implicit ordering. This keeps template composition debuggable: a fused spec can say exactly which bundle owns each segment.@@ -265,7 +265,7 @@ PPT Master uses **role switching within one main agent** rather than parallel su**Why role-specialized references, not one mega prompt.** Strategist runs in "negotiate with user" mode (open-ended, conversational, willing to back up); Executor runs in "produce strict XML" mode (no improvisation, no missing attributes). Mixing both into one prompt forces the model to hold incompatible discipline in the same turn — every prompt-engineering pathology of mode-mixing shows up. Splitting into per-role files lets each role load only what it needs and discard the rest.-**Eight Confirmations as the only blocking gate.** Strategist ends with eight bundled user confirmations (canvas / page count / audience / style / color / icon / typography / image). In the current workflow this is normally delivered through the Confirm UI in two tiers: Tier 1 confirms anchors (canvas, audience, divergence, delivery purpose, mode, visual style); Tier 2 is re-derived from the confirmed anchors and confirms realization choices (page count, palette, typography, icons, formula policy, image usage, AI image path, generation mode, refine-spec). The final `confirm_ui/result.json` is authoritative — if the user changes a field, the generated `design_spec.md` and `spec_lock.md` must use that value, not the AI's original recommendation.+**Strategist confirmation stage as the only blocking gate.** Strategist ends with a single confirmation gate delivered in three stages: Stage 1 confirms direction anchors (canvas, audience, divergence, delivery purpose, mode, visual style); Stage 2 is re-derived from those anchors and confirms the design system (page count, palette, typography, icons, formula policy); Stage 3 is re-derived from the confirmed design system and confirms images / execution (image usage, generated-image style, AI image path, generation mode, refine-spec). The final `confirm_ui/result.json` is authoritative — if the user changes a field, the generated `design_spec.md` and `spec_lock.md` must use that value, not the AI's original recommendation.**Image analysis goes through regenerated metadata, not pixels.** When images exist, Strategist and Executor use `analyze_images.py` output (`analysis/image_analysis.csv`) rather than directly opening image files. The CSV is a regenerated view over the live `images/` folder, not a durable cache. Re-running it before image-sensitive decisions is the staleness strategy: user images, extracted images, web images, AI outputs, formulas, and sliced elements all converge into the same measured fact table.@@ -318,7 +318,7 @@ Several architectural decisions shape this phase:**External refs during development, two divergent embedding strategies for delivery.** While editing in `svg_output/`, images are external file references — fast iteration, single-source-of-truth replacement. The two delivery artifacts then diverge: `svg_final/` Base64-inlines (a folder of self-contained SVGs that IDE preview, browser, and the preview pptx can all open without missing the bitmap dependencies); native pptx instead copies bitmaps into the PPTX media folder and uses `<a:srcRect>` to express the cropping. The split exists because Base64 inside DrawingML works but bloats file size 3-4×, while file-referenced bitmaps are PowerPoint's native idiom for which `<a:srcRect>` is the canonical crop expression — wrong tool in either direction would cost editability or file size.-**Three-dimensional AI image lock at Strategist time.** When the deck includes AI-generated images, Strategist decides three orthogonal dimensions up front — `rendering` (visual style family: vector-illustration / editorial / 3d-isometric / sketch-notes / …), `palette` (how the deck's HEX values are *used*: proportion + role + temperament), `type` (per-image internal composition: background / hero / framework / comparison / …). The first two are deck-wide and written into `spec_lock.md`; Image_Generator then assembles every per-image prompt from the single locked rendering + palette plus a per-image type, instead of re-deciding style per image. Without this, every image gets its own style drift and the deck reads as a stack of unrelated illustrations. This is the visual-cohesion dual of `spec_lock`'s typography/color anti-drift mechanism, just one level upstream of pixels. Strategist surfaces ≥3 candidate `rendering × palette` combinations to the user during the Eight Confirmations — never auto-locking a single combination silently, because the choice has far-reaching deck-wide consequences and the user's taste is the only oracle for it.+**Three-dimensional AI image lock at Strategist time.** When the deck includes AI-generated images, Strategist decides three orthogonal dimensions up front — `rendering` (visual style family: vector-illustration / editorial / 3d-isometric / sketch-notes / …), `palette` (how the deck's HEX values are *used*: proportion + role + temperament), `type` (per-image internal composition: background / hero / framework / comparison / …). The first two are deck-wide and written into `spec_lock.md`; Image_Generator then assembles every per-image prompt from the single locked rendering + palette plus a per-image type, instead of re-deciding style per image. Without this, every image gets its own style drift and the deck reads as a stack of unrelated illustrations. This is the visual-cohesion dual of `spec_lock`'s typography/color anti-drift mechanism, just one level upstream of pixels. Strategist surfaces ≥3 candidate `rendering × palette` combinations to the user during the Strategist confirmation stage — never auto-locking a single combination silently, because the choice has far-reaching deck-wide consequences and the user's taste is the only oracle for it.---@@ -473,7 +473,7 @@ Standalone workflows are route definitions, not optional decorations. They exist| `beautify-pptx` | existing PPTX, preserve page count/order/wording 1:1, improve layout | regenerate through SVG pipeline with source identity/content locked || `create-template` | build a reusable layout/deck template package | output a directory Step 3 can consume later || `create-brand` | extract or define a reusable brand identity | output `templates/brands/<id>/` |-| `resume-execute` | fresh chat after Phase A; user says to continue a project | enter Phase B without rerunning Strategist |+| `resume-execute` | fresh chat after the planning session; user says to continue a project | enter the execution session without rerunning Strategist || `refine-spec` | user explicitly wants to review/refine the spec before generation | stop after full spec/lock for revision, then resume || `verify-charts` | generated deck contains data charts | calibrate chart geometry before export || `customize-animations` | user asks for object-level animation order/effect/timing | create/validate `animations.json` and re-export policy |diff --git a/docs/templates-architecture.md b/docs/templates-architecture.mdindex 90a646bd..aa097681 100644--- a/docs/templates-architecture.md+++ b/docs/templates-architecture.md@@ -109,7 +109,7 @@ page_types: [<cover, toc, chapter, content, ending, ...>]| IV | Page Types | Per-page role (cover / toc / chapter / content / ending …) + variant descriptions || V | SVG Page Roster | File list + purpose, each file mapped to a III/IV role |-**Forbidden**: brand logo, brand voice & tone, official-truth color (`provenance: fact`) — those belong to brand. Layouts have no fallback color or typography by definition: identity segments are not written here; color and typography are decided live in Strategist's Eight Confirmations.+**Forbidden**: brand logo, brand voice & tone, official-truth color (`provenance: fact`) — those belong to brand. Layouts have no fallback color or typography by definition: identity segments are not written here; color and typography are decided live in Strategist's confirmation stage.### Deck schema@@ -207,7 +207,7 @@ When the user supplies a set of paths in their initial message, Step 3 fuses the|---|---|| (none) | Skip Step 3, free design || brand only | Copy brand wholesale; structure stays free design |-| layout only | Copy layout wholesale; identity stays free design (Strategist e/f/g confirmations decide) |+| layout only | Copy layout wholesale; identity stays free design (Strategist fields e/f/g decide) || deck only | Copy deck wholesale || brand + layout | brand provides identity, layout provides structure (follows existing SKILL.md fusion table) || brand + deck | brand overrides deck's identity segment at segment level; structure + middle come from deck |@@ -218,7 +218,7 @@ When the user supplies a set of paths in their initial message, Step 3 fuses theFusion defaults to **whole-segment integer replacement** — e.g. on deck + brand, the entire Color Scheme / Typography / Logo / Voice / Icon Style five sections come from brand. **No implicit field-level mixing** (you will never get "primary from brand, secondary from deck").-Field-level micro-adjustment goes through the existing Strategist Eight Confirmations path — the user says in chat "use the anthropic brand but change primary to #FF0000", and Strategist adjusts in confirmations e/g. Step 3 fusion does not add field-level syntax.+Field-level micro-adjustment goes through the existing Strategist confirmation stage path — the user says in chat "use the anthropic brand but change primary to #FF0000", and Strategist adjusts fields e/g. Step 3 fusion does not add field-level syntax.### Same-kind multiple paths = git-style conflict resolution@@ -270,9 +270,9 @@ This lets both AI and humans trace which segment came from where.| Multi-path | Fuse into one `design_spec.md` per the table above; merge SVG / logo files from each source || Same-kind multiple | Run the "git-style conflict resolution" prompt above to determine the merge |-### Strategist Eight Confirmations narrowing per kind+### Strategist confirmation stage narrowing per kind-When a deck path is supplied, the user already has a complete solution; the Eight Confirmations narrow to "target audience / page count / outline / tone tweaks" — deck-content fields. Other fields reuse the locked values directly. The narrowing rules live in `references/strategist.md` and `spec_lock_reference.md`.+When a deck path is supplied, the user already has a complete solution; the Strategist confirmation stage narrows to "target audience / page count / outline / tone tweaks" — deck-content fields. Other fields reuse the locked values directly. The narrowing rules live in `references/strategist.md` and `spec_lock_reference.md`.---@@ -289,6 +289,6 @@ After production, the frontmatter `kind` field determines whether the file lands## 7. Non-goals (rejection list paired with this framing)-- **No field-level override syntax in the fusion layer** — field-level adjustment uses the existing Strategist Eight Confirmations path+- **No field-level override syntax in the fusion layer** — field-level adjustment uses the existing Strategist confirmation stage path- **No batch conflict resolution for three or more of the same kind** — ask the user to narrow it down in chat first- **No bilingual name mapping table** — templates are named in their brand / scenario's native language (Chinese templates use Chinese names; English templates use snake_case); no forced unificationdiff --git a/docs/templates-guide.md b/docs/templates-guide.mdindex 64bae278..4e465a65 100644--- a/docs/templates-guide.md+++ b/docs/templates-guide.md@@ -59,7 +59,7 @@ A **style** is a description ("minimalist" / "Keynote-style" / "magazine 风")| | Template | Style ||---|---|---|| How invoked | Explicit directory path in your message | Free-form description in your message |-| What happens | Files copied into project; layouts inherit from template SVGs | Words flow to Strategist; color / typography / tone proposed in Eight Confirmations |+| What happens | Files copied into project; layouts inherit from template SVGs | Words flow to Strategist; color / typography / tone proposed in Strategist confirmation stage || Locked values | Yes — values come from the template's `design_spec.md` | No — Strategist invents values that fit the deck || Best for | Brand-locked decks; scenarios with strong visual conventions | When you have a feel in mind but no specific brand commitment |@@ -101,7 +101,7 @@ Three axes, freely combinable ("dark tech + minimalist" or "magazine + neo-Chine| **Memphis / pop** | High-saturation blocks, geometric, 80s || **Cyberpunk / vaporwave** | Neon purple-pink, grids, dreamlike |-When you describe a style, the AI doesn't pick a template — it interprets the words and lands them in Layer 2 of confirmation `d` (Style Objective) inside Strategist's Eight Confirmations, which then drives e (color), f (icon), g (typography), and h (image). You confirm or refine. If the style you want happens to match one of our built-in templates (e.g., `academic_defense` / `pixel_retro` / `psychology_attachment`), you have a choice: send the template's directory path for locked values, or describe the style for AI-interpreted values that adapt to your deck content.+When you describe a style, the AI doesn't pick a template — it interprets the words and lands them in Layer 2 of confirmation `d` (Style Objective) inside Strategist's confirmation stage, which then drives e (color), f (icon), g (typography), and h (image). You confirm or refine. If the style you want happens to match one of our built-in templates (e.g., `academic_defense` / `pixel_retro` / `psychology_attachment`), you have a choice: send the template's directory path for locked values, or describe the style for AI-interpreted values that adapt to your deck content.---diff --git a/docs/zh/faq.md b/docs/zh/faq.mdindex 6a6cb507..a9563669 100644--- a/docs/zh/faq.md+++ b/docs/zh/faq.md@@ -164,9 +164,9 @@ python3 skills/ppt-master/scripts/svg_to_pptx.py <project> -a auto --animation-t默认推荐**一次性连续生成**——10–15 页的 deck 在 200K 上下文窗口下完全够用,跨页视觉一致性也最好(Executor 看到前几页 SVG 后会主动对齐风格、字号、节奏)。-只有信号偏重的场景(页数 ≥ 18 / 源材料很厚 / 走过 topic-research 累积大量 web 抓取),AI 才会在策略师阶段给出**两段式(拆分模式)**的可选提示:第一阶段(八项确认 + 图片获取)结束后停止当前对话;你新开聊天窗口,输入 `继续生成 projects/<项目名>` 进入第二阶段(SVG 生成 + 导出)。新会话从磁盘重新加载 `design_spec` / `spec_lock` / `sources` / `images` 继续执行。+只有信号偏重的场景(页数 ≥ 18 / 源材料很厚 / 走过 topic-research 累积大量 web 抓取),AI 才会在策略师阶段给出**拆分模式**的可选提示:规划会话(策略师确认阶段 + 图片获取)结束后停止当前对话;你新开聊天窗口,输入 `继续生成 projects/<项目名>` 进入执行会话(SVG 生成 + 导出)。新会话从磁盘重新加载 `design_spec` / `spec_lock` / `sources` / `images` 继续执行。-两段式是**折中方案**——付出约 6K tokens 的 SKILL.md 重读成本,换得 60–200K 的 Phase A 噪声丢弃,并把节省下来的窗口空间用于 Phase B 主动重读 `sources/` 做内容增稠。**信号正常时不需要**,提示也不会出现;用户随时可以忽略提示,走默认连续模式。+两段式是**折中方案**——付出约 6K tokens 的 SKILL.md 重读成本,换得 60–200K 的规划会话噪声丢弃,并把节省下来的窗口空间用于执行会话主动重读 `sources/` 做内容增稠。**信号正常时不需要**,提示也不会出现;用户随时可以忽略提示,走默认连续模式。## Q: 能在导出前预览或修正某一页吗?diff --git a/docs/zh/roadmap.md b/docs/zh/roadmap.mdindex e237b455..ce8a9411 100644--- a/docs/zh/roadmap.md+++ b/docs/zh/roadmap.md@@ -52,8 +52,8 @@- **visual-style catalog 扩充至 18 个,与 image-renderings 对齐 + 示例库回收** — 先从[示例库](../../examples/)提炼 4 个(`brutalist` / `blueprint` / `memphis` / `zine`),再补齐 [`image-renderings`](../../skills/ppt-master/references/image-renderings/) 里有排版对应物的手绘 / 纹理风格 6 个(`sketch-notes` / `ink-notes` / `chalkboard` / `paper-cut` / `vintage-poster` / `pixel-art`),再回收示例库里仍未覆盖的独立气质:`ink-wash`(新中式水墨留白,源 藏拙 / 李子柒)· `glassmorphism`(深底磨砂玻璃 + 流光,源 glassmorphism_demo,从 soft-rounded 独立)· `photo-editorial`(满版摄影主导、文字点题,源 Pritzker / fashion_weekly)· `data-journalism`(Bloomberg/Economist 新闻信息图,多栏微图表 + 数据侧栏,源 global_ai_capital)。catalog 重组为 5 组(企业产品 / 编辑出版 / 表现印刷 / 手绘笔触 / 特殊)。**关键判据**:一个 rendering 升 visual-style 的前提是它定义「整页版面语言」而非「插入图的样子」——故 corporate-photo「摄影主导版面」该建(photo-editorial),而 nature / warm-scene / fantasy-animation 等纯氛围 rendering 仍只配对、不单建。全程零 HEX、Reference 强度- **mode catalog 扩档至 5 个:加 `briefing`** — 补上「中性信息平铺」这一格:无论点 / 无故事 / 不教学 / 不冲击,topic 标题、等权铺事实、完整可扫读,服务周报 / 参考册 / 目录 / 会议材料 / FAQ 这类「只告知不论证」的 deck。五个 mode 自此更接近 MECE 地切分**表达意图**:说服(pyramid)· 讲故事(narrative)· 教会(instructional)· 震住(showcase)· 只告知(briefing)。`_index` 加了 `briefing` vs `pyramid` 的灰区判据(「要不要造个 thesis 才塞得进 pyramid → 那就是 briefing」)。五个预设之外加一个 `custom` 兜底,承接预设盖不住的 bespoke 方向(特殊节奏 / 多 mode 融合 / 特定姿态)——用户点名**或策略师推荐**皆可,与所有锁一样由用户确认;一份 deck 永远只锁一个值,融合=一个 custom 描述多幕。唯一要避免的是「预设明明贴合却图省事甩 custom」。这与「用户自带大纲 / 方向覆盖 mode」是同一条真值优先原则- **mode / visual-style 体系真实 deck 验证完成 + 四项校准收紧落地** — 5 mode + 18 visual-style + `custom` 逃生舱在 5 份覆盖性 deck 上跑过验证(briefing×data-journalism / narrative×photo-editorial / instructional×chalkboard / showcase×glassmorphism / custom×zine,其中 narrative 一份走 AI 图生成分支):**选型零误判**(四对 Close-calls 灰区引力全被触发且全抗住)、**纪律全落实**(零 HEX / Reference 强度 / 整页版面语言)、**custom 机制可用**(`mode_behavior` 散文段落撑过 10 页生成、能讲成大白话让用户确认)、**mode ⟂ visual_style 正交成立**(任意组合无串味,含「keynote/发布会=mode 不是 style」路由正面验证)、导出 5/5 deck × 全页 0 失败。据真实信号收紧四处:`strategist §e` 按 visual_style 预判中性档位一次锁全(消除连续三份的 Executor 中途补色)、`executor-base §1` 套模板页重皮到当前 visual_style(模板供结构不供皮,镜像模板仍按 §1.1 逐字保留)、`briefing §1` 的 `core_message` = 本页覆盖什么而非证明什么(briefing 专属例外,全局 §IX 论断语义保留给 narrative/instructional/pyramid)、`svg_quality_checker` 修字体 drift 误报(按定界符匹配 + font-stack 归一化)+ 放宽 showcase mode 与 poster 类 visual-style 的字号上限-- **可选 spec 复核环节上线**([`workflows/refine-spec.md`](../../skills/ppt-master/workflows/refine-spec.md)) — 八项确认后新增一个 opt-in 停顿点:用户明确要求时(默认 OFF),Strategist 先产出完整 `design_spec.md` + `spec_lock.md`,停下来让用户对 spec 任意部分(大纲 / 配色 / 排版 / 版式 / 图片策略 / page rhythm)深入讨论修改,改完同步两个文件再进生成。与 split-mode 同构——不主动触发、默认管线零变化,仅在八项确认里多一行 opt-in 提示。复核视角(逻辑清晰度 / 信息密度 / 焦点 / 口语化 / 感染力 / 章节配比 + 各设计维度)只给方向、不落任何数字阈值(`Reference` 强度)。启发自 [@cuberoocp](https://github.com/cuberoocp) [issue #173](https://github.com/hugohe3/ppt-master/issues/173)-- **交互式可视化八项确认页(Step 4)**([`scripts/confirm_ui/server.py`](../../skills/ppt-master/scripts/confirm_ui/server.py),字段 schema [`scripts/docs/confirm_ui.md`](../../skills/ppt-master/scripts/docs/confirm_ui.md)) — 八项确认从纯聊天升级为浏览器可视化页面,Step 4 默认自动拉起:可枚举字段(canvas / mode / visual_style / 图标 / 图片用法 / 公式与生成策略)从 `catalogs.json` 出常用项、生成型字段摆候选——配色 swatch、字体实时预览(CJK / Latin 各自独立)、AI 图 rendering × palette 候选;并支持自定义 HEX 即时 swatch 反馈、配色 × 字体组合的实时合成预览、canvas 尺寸联动建议正文字号区间。与 Step 6 Live Preview **共用 `5050` 端口**(永不同时跑,确认页在 Step 4 末尾自动 `--shutdown` 让出端口;端口被占则自动顺延到下一个空闲口)。原则上**聊天始终是 canonical 通道、确认页只是便利层**:页面写出的 `result.json` 对推荐值具权威性、由下游就地消费(图片计划 / `image_strategy` / 字体 / split 与 refine-spec 开关都从中读取),任何打不开 / 超时 / 无 GUI 的情况都无损回退到聊天总结路径+- **可选 spec 复核环节上线**([`workflows/refine-spec.md`](../../skills/ppt-master/workflows/refine-spec.md)) — 策略师确认阶段后新增一个 opt-in 停顿点:用户明确要求时(默认 OFF),Strategist 先产出完整 `design_spec.md` + `spec_lock.md`,停下来让用户对 spec 任意部分(大纲 / 配色 / 排版 / 版式 / 图片策略 / page rhythm)深入讨论修改,改完同步两个文件再进生成。与 split-mode 同构——不主动触发、默认管线零变化,仅在策略师确认阶段里多一行 opt-in 提示。复核视角(逻辑清晰度 / 信息密度 / 焦点 / 口语化 / 感染力 / 章节配比 + 各设计维度)只给方向、不落任何数字阈值(`Reference` 强度)。启发自 [@cuberoocp](https://github.com/cuberoocp) [issue #173](https://github.com/hugohe3/ppt-master/issues/173)+- **交互式可视化策略师确认页(Step 4)**([`scripts/confirm_ui/server.py`](../../skills/ppt-master/scripts/confirm_ui/server.py),字段 schema [`scripts/docs/confirm_ui.md`](../../skills/ppt-master/scripts/docs/confirm_ui.md)) — 策略师确认阶段从纯聊天升级为浏览器可视化页面,Step 4 默认自动拉起:可枚举字段(canvas / mode / visual_style / 图标 / 图片用法 / 公式与生成策略)从 `catalogs.json` 出常用项、生成型字段摆候选——配色 swatch、字体实时预览(CJK / Latin 各自独立)、AI 图 rendering × palette 候选;并支持自定义 HEX 即时 swatch 反馈、配色 × 字体组合的实时合成预览、canvas 尺寸联动建议正文字号区间。与 Step 6 Live Preview **共用 `5050` 端口**(永不同时跑,确认页在 Step 4 末尾自动 `--shutdown` 让出端口;端口被占则自动顺延到下一个空闲口)。原则上**聊天始终是 canonical 通道、确认页只是便利层**:页面写出的 `result.json` 对推荐值具权威性、由下游就地消费(图片计划 / `image_strategy` / 字体 / split 与 refine-spec 开关都从中读取),任何打不开 / 超时 / 无 GUI 的情况都无损回退到聊天总结路径- **源文档转换保真度提升一批** — 源材料进管线时更少丢信息:`doc_to_md` 把 Word 里的 OMML / Office Math 公式转成内联 LaTeX、`pdf_to_md` 识别 `Figure N |` 竖线分隔的图注、`ppt_to_md` 保留源 deck 已有的超链接(run 级外链 `[text](url)` / slide 内部跳转 `[text](#slide-N)` / shape 级点击,含危险 scheme 过滤与锚文本 Markdown 转义)并把原生图表数据转写成 Markdown 表格(数值随转换存活,不再只剩一张图)。图注识别基于 [@suay1113](https://github.com/suay1113) [PR #191](https://github.com/hugohe3/ppt-master/pull/191),超链接保留提炼自 [@ZhaoZuohong](https://github.com/ZhaoZuohong) [PR #155](https://github.com/hugohe3/ppt-master/pull/155)- **内容保真的 PPT 美化 / 重排版上线**([`workflows/beautify-pptx.md`](../../skills/ppt-master/workflows/beautify-pptx.md)) — 与 `template-fill` 互为镜像:template-fill 复用某份 deck 的设计换新内容,beautify 反过来保留内容、重做版面。给一份现成 PPTX,**全部文字逐字保留(不增 / 不删 / 不改写)**,从源 deck 提取并**继承其视觉身份(配色 / 字体,`theme` 或 `observed` 两套候选过确认页)**,只重做版面 / 层级 / 留白;严格 1:1 页数页序,图表 / 表格从抽取数据原生重绘(数据冻结)、源配图重新排布。技术上仍走从零生成原生 PPTX 管线(`ppt_to_md` 抽内容 → 主管线 → 全新 deck),不补丁原文件,因此不碰 Non-goals #53。新增 `beautify_identity.py` / `beautify_inventory.py`,confirm 页全字段按源 seed 后用户复核。v1 天花板(诚实标注):不缓解信息过载(挤页只在页内改,真要重排分页属主管线)、不保证坐标级 paste-back、combo / dual-axis / waterfall 图丢未捕获的绘图层diff --git a/docs/zh/technical-design.md b/docs/zh/technical-design.mdindex 05a3d3bb..943a6404 100644--- a/docs/zh/technical-design.md+++ b/docs/zh/technical-design.md@@ -27,7 +27,7 @@仅在用户提供明确模板目录路径且其中 design_spec.md 声明 kind: brand/layout/deck 时触发原生 PPTX 模板请求进入 template-fill;可复用 SVG 模板需先通过 create-template 创建↓-[Strategist] 策略师 - 两层八项确认与设计规范 → design_spec.md + spec_lock.md+[Strategist] 策略师 - 三阶段策略师确认与设计规范 → design_spec.md + spec_lock.md↓[Image Acquisition] 图片获取(当资源列表中有需要 AI 生成、网络搜索或切片的图片时)↓@@ -246,7 +246,7 @@ PPT Master 不只服务 PPT——同一套 SVG → DrawingML 流水线还能产| Kind | 拥有的片段 | 典型内容 | 对 Strategist 的影响 ||---|---|---|---|| `brand` | 身份片段 | 配色、字体、logo、语气、图标风格 | 锁定身份;结构保持自由 |-| `layout` | 结构片段 | 画布、页面结构、页面类型、SVG roster | 锁定结构;身份仍在八项确认里确定 |+| `layout` | 结构片段 | 画布、页面结构、页面类型、SVG roster | 锁定结构;身份仍在策略师确认阶段里确定 || `deck` | 身份 + 结构 + 模板总览 | 完整复刻型包 | 锁定完整模板语法,只剩内容相关选择 |当用户提供多个路径时,融合是**片段级**而不是字段级:brand 覆盖身份片段,layout 覆盖结构片段,deck 提供中间的 template overview 片段。同类冲突会被显式列为冲突,而不是按输入顺序默默决定。这样融合后的 spec 能明确说明每个片段来自哪里,便于审计和复现。@@ -265,7 +265,7 @@ PPT Master 用的是**单主代理内的角色切换**,不是并行子代理**为什么是角色专属 reference 而不是一个超大 prompt。** Strategist 跑的是「跟用户协商」模式(开放式、对话式、可以回退),Executor 跑的是「产出严格 XML」模式(不准即兴、不准漏属性)。把两者塞进同一个 prompt,强迫模型在同一个 turn 里持守相互矛盾的纪律——所有混合模式的 prompt 工程病灶都会出现。按角色拆开,每个角色只加载它需要的、扔掉其他。-**Eight Confirmations 是唯一的阻塞 gate。** Strategist 阶段以八项打包确认(画布 / 页数 / 受众 / 风格 / 配色 / 图标 / 排版 / 图像)作为核心阻塞决策点。当前流程默认通过 Confirm UI 分两层呈现:Tier 1 确认锚点(画布、受众、改写幅度、交付目的、mode、visual style);Tier 2 基于用户已确认的锚点重新推导,再确认实现层选择(页数、调色板、字体、图标、公式策略、图片来源、AI 图片路径、生成模式、refine-spec)。最终的 `confirm_ui/result.json` 是权威输入——用户改过的字段必须进入 `design_spec.md` 和 `spec_lock.md`,不能回退到 AI 的原始推荐。+**策略师确认阶段是唯一的阻塞 gate。** Strategist 阶段以一个三阶段确认 gate 作为核心阻塞决策点:第一阶段确认方向锚点(画布、受众、改写幅度、交付目的、mode、visual style);第二阶段基于已确认锚点重新推导并确认设计系统(页数、调色板、字体、图标、公式策略);第三阶段基于已确认设计系统重新推导并确认图片与执行方式(图片来源、生成图风格、AI 图片路径、生成模式、refine-spec)。最终的 `confirm_ui/result.json` 是权威输入——用户改过的字段必须进入 `design_spec.md` 和 `spec_lock.md`,不能回退到 AI 的原始推荐。**图片分析走重算元数据,不读像素。** 当项目里存在图片时,Strategist 和 Executor 使用 `analyze_images.py` 的输出(`analysis/image_analysis.csv`),而不是直接打开图片文件。这个 CSV 是基于当前 `images/` 目录重算出来的视图,不是持久缓存。每次做图片敏感决策前重跑分析,就是它的防陈旧策略:用户图、抽取图、网络图、AI 图、公式图和切片图最终都会汇入同一张可度量事实表。@@ -318,7 +318,7 @@ Strategist 阶段产出两份看起来冗余但服务不同对象的产物:**开发期外部引用,交付期分叉成两套嵌入策略。** 在 `svg_output/` 里编辑时,图片是外部文件引用——快速迭代、单点替换。两份交付产物随后分叉:`svg_final/` 走 Base64 内联(产出一组自包含 SVG,IDE 预览、浏览器、preview pptx 都能开而不丢位图依赖);native pptx 反过来把位图复制进 PPTX 的 media 文件夹,用 `<a:srcRect>` 表达裁剪。分叉的理由:在 DrawingML 里塞 Base64 能跑但文件膨胀 3-4 倍;文件引用的位图是 PowerPoint 原生表达方式,配 `<a:srcRect>` 的裁剪也是 DrawingML 的规范写法——任一方向用错工具都要付出可编辑性或文件大小的代价。-**AI 图片三维系统:Strategist 阶段就锁定。** 当 deck 包含 AI 生成图片时,Strategist 在前置阶段一次性确定三个正交维度——`rendering`(视觉风格家族:vector-illustration / editorial / 3d-isometric / sketch-notes / ……)、`palette`(deck 的 HEX 在图里**怎么用**:比例 + 角色 + 气质)、`type`(每张图的内部构图:background / hero / framework / comparison / ……)。前两个是 deck 级、写进 `spec_lock.md`;Image_Generator 此后每张图的 prompt 都从同一份锁定的 rendering + palette 加上该图的 type 组装出来,而不是逐图重决风格。没有这层锁定,每张图都会自己风格漂移,整套 deck 读起来就是一摞互不相关的插画。这是 `spec_lock` 字体/色彩抗漂移机制在像素上游的对偶——同一思路,往前推一层。Strategist 在八项确认阶段会向用户呈现 **≥3 个 `rendering × palette` 候选**,绝不静默地自动锁定单一组合,因为这是一个会牵动全 deck 视觉的选择,唯一权威只有用户的品味。+**AI 图片三维系统:Strategist 阶段就锁定。** 当 deck 包含 AI 生成图片时,Strategist 在前置阶段一次性确定三个正交维度——`rendering`(视觉风格家族:vector-illustration / editorial / 3d-isometric / sketch-notes / ……)、`palette`(deck 的 HEX 在图里**怎么用**:比例 + 角色 + 气质)、`type`(每张图的内部构图:background / hero / framework / comparison / ……)。前两个是 deck 级、写进 `spec_lock.md`;Image_Generator 此后每张图的 prompt 都从同一份锁定的 rendering + palette 加上该图的 type 组装出来,而不是逐图重决风格。没有这层锁定,每张图都会自己风格漂移,整套 deck 读起来就是一摞互不相关的插画。这是 `spec_lock` 字体/色彩抗漂移机制在像素上游的对偶——同一思路,往前推一层。Strategist 在策略师确认阶段会向用户呈现 **≥3 个 `rendering × palette` 候选**,绝不静默地自动锁定单一组合,因为这是一个会牵动全 deck 视觉的选择,唯一权威只有用户的品味。---@@ -473,7 +473,7 @@ PowerPoint 的 DrawingML 是 SVG 表达力的严格子集。Executor 在一份| `beautify-pptx` | 现有 PPTX,页数/页序/措辞必须 1:1 保留,只改善排版 | 锁定源身份与内容后,通过 SVG 流水线重新生成 || `create-template` | 构建可复用 layout/deck 模板包 | 输出后续 Step 3 可消费的目录 || `create-brand` | 提取或定义可复用品牌身份 | 输出 `templates/brands/<id>/` |-| `resume-execute` | Phase A 后新开聊天,用户要求继续某项目 | 不重跑 Strategist,直接进入 Phase B |+| `resume-execute` | 规划会话后新开聊天,用户要求继续某项目 | 不重跑 Strategist,直接进入执行会话 || `refine-spec` | 用户明确要求生成前先审阅 / 修改 spec | 写出完整 spec/lock 后停下,用户修改后再恢复 || `verify-charts` | 生成 deck 含数据图表 | 导出前校准图表几何 || `customize-animations` | 用户要求调对象级动画顺序 / 效果 / 计时 | 创建 / 校验 `animations.json` 并控制再导出策略 |diff --git a/docs/zh/templates-architecture.md b/docs/zh/templates-architecture.mdindex 6c0024ad..01531046 100644--- a/docs/zh/templates-architecture.md+++ b/docs/zh/templates-architecture.md@@ -109,7 +109,7 @@ page_types: [<cover, toc, chapter, content, ending, ...>]| IV | Page Types | 每种页面的角色(cover / toc / chapter / content / ending …)与变体说明 || V | SVG Page Roster | 文件清单 + 用途,每个文件对应 III/IV 哪一类 |-**不允许出现**:品牌 logo、品牌 voice & tone、官方真值色(`provenance: fact`)——这些是 brand 的职责。Layout 自身没有兜底色/字体(这是定义:layout 不写身份段;色彩与字体在 Strategist 八项确认现场决策)。+**不允许出现**:品牌 logo、品牌 voice & tone、官方真值色(`provenance: fact`)——这些是 brand 的职责。Layout 自身没有兜底色/字体(这是定义:layout 不写身份段;色彩与字体在 策略师确认阶段现场决策)。### Deck schema@@ -207,7 +207,7 @@ primary_color: "<HEX>"|---|---|| 无 | 跳过 Step 3,走自由设计 || 只 brand | 复制 brand 全部,结构走自由设计 |-| 只 layout | 复制 layout 全部,身份走自由设计(Strategist 八项确认 e/f/g 决策) |+| 只 layout | 复制 layout 全部,身份走自由设计(策略师确认阶段 e/f/g 决策) || 只 deck | 复制 deck 全部 || brand + layout | brand 提供身份段 + layout 提供结构段,沿用 SKILL.md 现有 fusion 表 || brand + deck | brand 段级覆盖 deck 的身份段,结构段与中间段从 deck 拿 |@@ -218,7 +218,7 @@ primary_color: "<HEX>"合成默认是**段级整段替换**——例如 deck + brand 时,整个 Color Scheme / Typography / Logo / Voice / Icon Style 五段从 brand 拿,**不做字段级混搭**(即不会发生"primary 从 brand 拿、secondary 从 deck 拿"这类隐式混合)。-字段级微调走 Strategist 八项确认这条已有路径——用户在 chat 里说"用 anthropic brand,但 primary 改成 #FF0000",由 Strategist 在 e/g 现场调整,不在 Step 3 的 fusion 层加字段级语法。+字段级微调走 策略师确认阶段这条已有路径——用户在 chat 里说"用 anthropic brand,但 primary 改成 #FF0000",由 Strategist 在 e/g 现场调整,不在 Step 3 的 fusion 层加字段级语法。### 同类多份 = git 冲突解决@@ -270,9 +270,9 @@ AI: 你给了两个 brand,检测到段级冲突:> 位图统一进项目 `images/`(和 AI / 网络 / 用户图片同一个运行期图片池,SVG 里走 `../images/`);`templates/` 只放 spec 和模板 SVG 等供 Strategist/Executor 阅读、不被直接渲染的参考材料。| 同类多份 | 按上节"git 冲突解决"问答,得到合成结果 |-### Strategist 八项确认在不同 kind 下的收窄+### 策略师确认阶段在不同 kind 下的收窄-Deck 路径下用户已经拿到完整方案,八项确认收窄到"目标受众 / 页数 / 大纲 / 调性微调"等 deck 内容相关字段;其他字段直接从锁定值复用。具体收窄规则落在 `references/strategist.md` 与 `spec_lock_reference.md`。+Deck 路径下用户已经拿到完整方案,策略师确认阶段收窄到"目标受众 / 页数 / 大纲 / 调性微调"等 deck 内容相关字段;其他字段直接从锁定值复用。具体收窄规则落在 `references/strategist.md` 与 `spec_lock_reference.md`。---@@ -289,6 +289,6 @@ Deck 路径下用户已经拿到完整方案,八项确认收窄到"目标受## 七、不做(与本文 framing 配套的拒绝列表)-- **不在 fusion 层支持字段级覆盖语法** —— 字段级微调走 Strategist 八项确认这条已有路径+- **不在 fusion 层支持字段级覆盖语法** —— 字段级微调走 策略师确认阶段这条已有路径- **不为同类三份及以上设计批量冲突解决** —— 用户先在 chat 里收敛到两份- **不引入双名映射表** —— 模板命名按其品牌/场景母语(中文模板用中文名,英文模板用 snake_case),不强制统一diff --git a/docs/zh/templates-guide.md b/docs/zh/templates-guide.mdindex 5fac061b..320b05ad 100644--- a/docs/zh/templates-guide.md+++ b/docs/zh/templates-guide.md@@ -59,7 +59,7 @@ AI 会把这个目录里的 SVG、`design_spec.md` 和素材复制到项目目| | 模板 | 风格 ||---|---|---|| 怎么触发 | 消息里给出明确的目录路径 | 消息里写自由描述 |-| 发生什么 | 文件复制到项目;layouts 继承自模板 SVG | 描述流到 Strategist;色彩 / 字体 / 调性在八项确认里推荐 |+| 发生什么 | 文件复制到项目;layouts 继承自模板 SVG | 描述流到 Strategist;色彩 / 字体 / 调性在策略师确认阶段里推荐 || 数值锁定 | 是 — 来源于模板的 `design_spec.md` | 否 — Strategist 现场推适合 deck 的具体值 || 适用场景 | 品牌锁定的 deck;强视觉约定的场景 | 心里有感觉但没有具体品牌承诺 |@@ -101,7 +101,7 @@ AI 会把这个目录里的 SVG、`design_spec.md` 和素材复制到项目目| **孟菲斯/波普风** | 高饱和大色块、几何图形、80 年代 || **赛博朋克/蒸汽波** | 霓虹紫粉、网格、迷幻 |-你描述风格时,AI **不会基于这些词去挑模板**——它把这些词解释为对应的色彩 / 字体 / 版式建议,放到 Strategist 八项确认里 `d` 项的第二层(视觉风格),然后驱动 e/f/g/h(色彩 / 图标 / 字体 / 图片)。你可以确认或调整。如果你想要的风格刚好对上库里某个模板(如 `academic_defense` / `pixel_retro` / `psychology_attachment`),有两条路可选:把模板的目录路径发出来锁定值,或描述风格让 AI 现场推适配你内容的值。+你描述风格时,AI **不会基于这些词去挑模板**——它把这些词解释为对应的色彩 / 字体 / 版式建议,放到 策略师确认阶段里 `d` 项的第二层(视觉风格),然后驱动 e/f/g/h(色彩 / 图标 / 字体 / 图片)。你可以确认或调整。如果你想要的风格刚好对上库里某个模板(如 `academic_defense` / `pixel_retro` / `psychology_attachment`),有两条路可选:把模板的目录路径发出来锁定值,或描述风格让 AI 现场推适配你内容的值。---diff --git a/skills/ppt-master/SKILL.md b/skills/ppt-master/SKILL.mdindex 34f7b277..5db12f30 100644--- a/skills/ppt-master/SKILL.md+++ b/skills/ppt-master/SKILL.md@@ -20,7 +20,7 @@ description: >>> 1. **SERIAL EXECUTION** — Steps MUST be executed in order; the output of each step is the input for the next. Non-BLOCKING adjacent steps may proceed continuously once prerequisites are met, without waiting for the user to say "continue"> 2. **BLOCKING = HARD STOP** — Steps marked ⛔ BLOCKING require a full stop; the AI MUST wait for an explicit user response before proceeding and MUST NOT make any decisions on behalf of the user-> 3. **NO CROSS-PHASE BUNDLING** — Cross-phase bundling is FORBIDDEN. (Note: the Eight Confirmations in Step 4 are ⛔ BLOCKING — the AI MUST present recommendations and wait for explicit user confirmation before proceeding. Once the user confirms, all subsequent non-BLOCKING steps — design spec output, SVG generation, speaker notes, and post-processing — may proceed automatically without further user confirmation)+> 3. **NO CROSS-PHASE BUNDLING** — Cross-phase bundling is FORBIDDEN. (Note: the Strategist confirmation stage in Step 4 is ⛔ BLOCKING — the AI MUST present recommendations and wait for explicit user confirmation before proceeding. Once the user confirms, all subsequent non-BLOCKING steps — design spec output, SVG generation, speaker notes, and post-processing — may proceed automatically without further user confirmation)> 4. **GATE BEFORE ENTRY** — Each Step has prerequisites (🚧 GATE) listed at the top; these MUST be verified before starting that Step> 5. **NO SPECULATIVE EXECUTION** — "Pre-preparing" content for subsequent Steps is FORBIDDEN (e.g., writing SVG code during the Strategist phase)> 6. **NO SUB-AGENT SVG GENERATION** — Executor Step 6 SVG generation is context-dependent and MUST be completed by the current main agent end-to-end. Delegating page SVG generation to sub-agents is FORBIDDEN@@ -218,7 +218,7 @@ Do **not** reinterpret this boundary as 1:1 redesign or free SVG generation. UseThere is no slug matching, no name lookup, no fuzzy resolution. A name without a path does not trigger — the user must give a path the AI can `cd` into.-> Style descriptions ("麦肯锡风格" / "Keynote 风" / "极简风" / etc.) never trigger Step 3. They flow into Strategist's Eight Confirmations as a style brief (color / typography / tone in confirmations e–g).+> Style descriptions ("麦肯锡风格" / "Keynote 风" / "极简风" / etc.) never trigger Step 3. They flow into the Strategist confirmation stage as a style brief (color / typography / tone in fields e–g).> Bare names ("academic_defense", "招商银行", "anthropic") do NOT trigger Step 3 even if a matching directory exists in the library. The user must give a path. AI must not "helpfully" resolve a name to a path.@@ -249,8 +249,8 @@ The architecture has three independent reference bundles. Full schema in [`docs/| User path's `kind` | Step 3 action ||---|---|| `kind: brand` | `design_spec.md` + non-image assets → `<project>/templates/`; logo / illustration / icon **bitmaps** → `<project>/images/`. Strategist locks identity segment as truth; structure stays free. |-| `kind: layout` | `design_spec.md` + SVG roster → `<project>/templates/`; any **bitmap** assets → `<project>/images/`. Strategist locks structure; identity decided in Eight Confirmations e–g. |-| `kind: deck` | `design_spec.md` + template SVGs → `<project>/templates/`; logos / backgrounds / other **bitmaps** → `<project>/images/`. Strategist locks all segments; Eight Confirmations narrows to deck-content fields (audience / page count / outline / tone tweaks). |+| `kind: layout` | `design_spec.md` + SVG roster → `<project>/templates/`; any **bitmap** assets → `<project>/images/`. Strategist locks structure; identity decided in Strategist confirmation stage e–g. |+| `kind: deck` | `design_spec.md` + template SVGs → `<project>/templates/`; logos / backgrounds / other **bitmaps** → `<project>/images/`. Strategist locks all segments; Strategist confirmation stage narrows to deck-content fields (audience / page count / outline / tone tweaks). |```bashTEMPLATE_DIR=<user-supplied path>@@ -279,7 +279,7 @@ Override priority by segment:| layout + deck | deck | layout (overrides deck) | deck || brand + layout + deck | brand | layout | deck |-Field-level micro-adjustment (e.g. "use anthropic brand but primary changed to #FF0000") is **not** part of Step 3 fusion — it flows into Strategist Eight Confirmations e–g as a normal user request.+Field-level micro-adjustment (e.g. "use anthropic brand but primary changed to #FF0000") is **not** part of Step 3 fusion — it flows into Strategist confirmation stage e–g as a normal user request.#### Same-kind multiple paths — conflict resolution@@ -333,13 +333,13 @@ Read references/strategist.md**Artifact ownership**: fact-channel and source/derived artifact boundaries are defined in [`references/artifact-ownership.md`](references/artifact-ownership.md). This Step uses those ownership rules; it does not redefine them.-**`<project_path>/analysis/` is the project's intermediate-analysis folder: the canonical home for machine-extracted source/asset facts — the PPTX intake bundle (`source_profile.json` index + per-deck `<stem>.identity.json` / `<stem>.slide_library.json`) and `image_analysis.csv`. It holds facts, not design contracts — `design_spec.md` / `spec_lock.md` stay at the project root.** The MUST-read contract covers only the **compact structured data files (`.json` / `.csv`)**; other artifacts that may live under `analysis/` (e.g. a beautify `source_svg_import/` vector reference package) are NOT bulk-read — they are read selectively only when a specific workflow step calls for them. Before the Eight Confirmations, Strategist MUST read the auto-extracted fact files already in `analysis/` — currently `source_profile.json` (PPTX intake), when present. This file is the multi-deck index: read it once for the `decks[]` digests (canvas / chart / table entries per source deck), then open a specific deck's `<stem>.identity.json` / `<stem>.slide_library.json` only if you need its full raw facts. Use these entries as **factual source context** (format default + content facts); when several decks are present, synthesize across all of them. The source's **palette / typography / visual identity are a reference, not a constraint**: the main pipeline may inherit them where they fit the content and the confirmed style, or design fresh where they don't — the Strategist's judgment, never an obligation to either keep or discard. (Template-fill preserves the native source design by editing cloned slides directly; beautify defaults to the source identity but still follows the confirmed values; the main pipeline treats source identity as reference only and defaults to fresh design.) (`image_analysis.csv` lands later, at the image-analysis step below, and is the authoritative regenerated image-fact view there — re-derived from the live `images/` folder, not a durable store.)+**`<project_path>/analysis/` is the project's intermediate-analysis folder: the canonical home for machine-extracted source/asset facts — the PPTX intake bundle (`source_profile.json` index + per-deck `<stem>.identity.json` / `<stem>.slide_library.json`) and `image_analysis.csv`. It holds facts, not design contracts — `design_spec.md` / `spec_lock.md` stay at the project root.** The MUST-read contract covers only the **compact structured data files (`.json` / `.csv`)**; other artifacts that may live under `analysis/` (e.g. a beautify `source_svg_import/` vector reference package) are NOT bulk-read — they are read selectively only when a specific workflow step calls for them. Before the Strategist confirmation stage, Strategist MUST read the auto-extracted fact files already in `analysis/` — currently `source_profile.json` (PPTX intake), when present. This file is the multi-deck index: read it once for the `decks[]` digests (canvas / chart / table entries per source deck), then open a specific deck's `<stem>.identity.json` / `<stem>.slide_library.json` only if you need its full raw facts. Use these entries as **factual source context** (format default + content facts); when several decks are present, synthesize across all of them. The source's **palette / typography / visual identity are a reference, not a constraint**: the main pipeline may inherit them where they fit the content and the confirmed style, or design fresh where they don't — the Strategist's judgment, never an obligation to either keep or discard. (Template-fill preserves the native source design by editing cloned slides directly; beautify defaults to the source identity but still follows the confirmed values; the main pipeline treats source identity as reference only and defaults to fresh design.) (`image_analysis.csv` lands later, at the image-analysis step below, and is the authoritative regenerated image-fact view there — re-derived from the live `images/` folder, not a durable store.)**Channel ownership — read each fact once from its owning channel.** In the main pipeline the **content contract is the content-type files in `sources/`** — primarily `<stem>.md`, but also any user-supplied content the import archived there: `.md` / `.markdown` / `.txt` / `.csv` / `.tsv` / `.json` / `.jsonl` / `.yaml` / `.yml` (a `metrics.json` or `data.csv` may carry core content — judge by what the file holds). Text, tables, and chart data values come from these (`ppt_to_md` now transcribes native chart data into Markdown tables). **Do NOT read pipeline sidecars in `sources/` as content**: `*.conversion_profile.json` (conversion audit) and `*_files/image_manifest.json` (asset index) are process metadata — open them only to audit a conversion or resolve assets, never as slide content. Converted-source originals archived in `sources/` (`.pdf` / `.pptx` / `.docx` / `.xlsx` / `.html` / `.epub` / `.tex` / `.rst` / `.ipynb` / `.typ`, etc.) are read via their converted `<stem>.md`, not scanned directly in the main pipeline. The `analysis/` chart / table entries are a **structural digest** for outline decisions (which slides carried charts, type, series names) — not a second copy of the values; do NOT also pull chart values from `<stem>.slide_library.json` in the main pipeline. The `<stem>.slide_library.json` full structured data is owned by the direct-PPTX workflows: template-fill uses it as the native fill contract; beautify uses it for native chart / table data while keeping slide text from the Markdown.-**Eight Confirmations** (full template: `templates/design_spec_reference.md`):+**Strategist confirmation stage** (full template: `templates/design_spec_reference.md`):-⛔ **BLOCKING**: present the Eight Confirmations and **wait for explicit user confirmation or modification** before outputting Design Specification & Content Outline. This is the single core confirmation gate — once the final confirmation lands, all subsequent steps proceed automatically. The default Confirm UI delivers the gate in **three stages** (direction → design system → images / execution; see below); the chat fallback mirrors the same staged order.+⛔ **BLOCKING**: present the Strategist confirmation stage and **wait for explicit user confirmation or modification** before outputting Design Specification & Content Outline. This is the single core confirmation gate — once the final confirmation lands, all subsequent steps proceed automatically. The default Confirm UI delivers the gate in **three stages** (direction → design system → images / execution; see below); the chat fallback mirrors the same staged order.1. Canvas format2. Page count range@@ -350,7 +350,7 @@ Read references/strategist.md7. Typography plan, including formula rendering policy8. Image usage approach-**Confirm UI Auto-Launch (Mandatory — default visual confirmation surface)**: by default the Eight Confirmations are presented through an interactive local page in **three stages within one browser session** — Stage 1 confirms the direction anchors; the AI then re-derives the design-system layer from the **user's actual** anchors; Stage 2 confirms that layer; the AI then re-derives image and execution choices from the confirmed direction + design system; Stage 3 confirms the final operational layer. Color swatches, live font previews, icon samples, image-style reference previews, and candidate picks appear where they help judgment; the chat path is the always-valid fallback. [`scripts/docs/confirm_ui.md`](scripts/docs/confirm_ui.md) owns the schema, server lifecycle, port strategy, and fallback details; this section keeps the orchestration contract. The split:+**Confirm UI Auto-Launch (Mandatory — default visual confirmation surface)**: by default the Strategist confirmation stage is presented through an interactive local page in **three stages within one browser session** — Stage 1 confirms the direction anchors; the AI then re-derives the design-system layer from the **user's actual** anchors; Stage 2 confirms that layer; the AI then re-derives image and execution choices from the confirmed direction + design system; Stage 3 confirms the final operational layer. Color swatches, live font previews, icon samples, image-style reference previews, and candidate picks appear where they help judgment; the chat path is the always-valid fallback. [`scripts/docs/confirm_ui.md`](scripts/docs/confirm_ui.md) owns the schema, server lifecycle, port strategy, and fallback details; this section keeps the orchestration contract. The split:| Stage | Confirms | Driven by ||---|---|---|@@ -364,18 +364,18 @@ Steps:> ⛔ **Steps 2 → 3 → 4 are ONE uninterrupted run — do NOT yield to the user mid-flow.** When an intermediate `--wait` returns, the AI **immediately and autonomously** re-derives and writes the next stage in the **same turn**: do **not** summarize, ask a question, report progress, or end the turn in between. The browser is sitting on a "deriving…" spinner polling for the next stage you must write — stopping here strands the page and the user must prod you in chat to finish (a bug, not the intended flow). **Stage-1 and Stage-2 confirmations are intermediate machine handoffs, not stopping points.** The single ⛔ BLOCKING wait is the **final** confirmation at the end of step 4. (Chat-fallback path — only when the page never opened — is the exception: there you do present each stage in chat and wait for a reply.)-1. **Write Stage 1** to `<project_path>/confirm_ui/recommendations.json` with `"tier": 1` and only the anchor fields. `tier` is the internal JSON stage selector retained for compatibility; user-facing documentation and UI call these stages. Enumerable anchors (`canvas` / `mode` / `visual_style` / `delivery_purpose`) name a recommended canonical `id` in a `recommend` block (the page lists common options from `confirm_ui/static/catalogs.json`); `visual_style` also carries the ≥3-style `visual_style_spectrum` (safe / shifted / bold — same hard rule as h.5). `audience` and `content_divergence` are plain `{ "value": "<free text>" }`. `content_divergence` is the **free-text** field shown under audience in §c — how closely to follow the source vs how freely to reshape it (blank = balanced; facts stay sourced at every level); it is consumed by Strategist when authoring `§IX`, recorded in `design_spec.md §I`, carries no page-count coupling, and is **not** written to `spec_lock.md`. Set `lang` to the page language (`zh` / `en` / `ja`); visible text matches `lang`, or provide multilingual `name_zh` / `name_en` / `name_ja` + `note_zh` / `note_en` / `note_ja` — when the user's language is Japanese, set `lang: "ja"` and always include the `_ja` variants (labels resolve in the page language first — a `ja` page falls back ja → en → zh, so missing `_ja` labels silently render in English; zh/en pages keep their zh↔en fallback and only try `_ja` last).+1. **Write Stage 1** to `<project_path>/confirm_ui/recommendations.json` with `"stage": "stage1"` and only the anchor fields. New recommendations MUST use the canonical `stage` selector. Enumerable anchors (`canvas` / `mode` / `visual_style` / `delivery_purpose`) name a recommended canonical `id` in a `recommend` block (the page lists common options from `confirm_ui/static/catalogs.json`); `visual_style` also carries the ≥3-style `visual_style_spectrum` (safe / shifted / bold — same hard rule as h.5). `audience` and `content_divergence` are plain `{ "value": "<free text>" }`. `content_divergence` is the **free-text** field shown under audience in §c — how closely to follow the source vs how freely to reshape it (blank = balanced; facts stay sourced at every level); it is consumed by Strategist when authoring `§IX`, recorded in `design_spec.md §I`, carries no page-count coupling, and is **not** written to `spec_lock.md`. Set `lang` to the page language (`zh` / `en` / `ja`); visible text matches `lang`, or provide multilingual `name_zh` / `name_en` / `name_ja` + `note_zh` / `note_en` / `note_ja` — when the user's language is Japanese, set `lang: "ja"` and always include the `_ja` variants (labels resolve in the page language first — a `ja` page falls back ja → en → zh, so missing `_ja` labels silently render in English; zh/en pages keep their zh↔en fallback and only try `_ja` last).2. **Launch + wait for Stage 1.** Background launch; the parent returns when the page writes the stage-1 `result.json`. **Long tool timeout — 600000 ms** (the `--wait` ≈590 s budget):```bashpython3 ${SKILL_DIR}/scripts/confirm_ui/server.py <project_path> --daemon --wait```- Page opens at `http://localhost:5050` — the **same port as the Step 6 live preview** (they never run at once: this page shuts down at the end of Step 4). If 5050 is held, the launcher **auto-advances** (5051, …) — read the actual URL from the launch log and report it. The page does **not** close after Stage 1: it shows a "deriving…" state and polls for Stage 2. **Launch or wait failure is non-fatal**: if it fails or times out (flask missing, port blocked, no GUI / remote / web host), do **NOT** troubleshoot — **on any non-zero exit, re-check `result.json` once** (a fresh `status: tier1-confirmed`) before dropping to the chat fallback. **On success (exit 0 with a stage-1 result), do not pause or report — go straight to step 3 in the same turn.**-3. **Re-derive Stage 2 from the confirmed anchors, write it, then wait for the design-system handoff — immediately, same turn (the page is polling for it).** Read the stage-1 `result.json` (`status: tier1-confirmed`). Using the user's **actual** confirmed anchors (not your originals), author the design-system candidates and **overwrite** `recommendations.json` with `"tier": 2`: page count (content volume × `delivery_purpose`); color and typography as **generative ≥3-candidate** fields (creative recommendations always offer real choice; fewer than 3 only on the honest-shortfall exception, with a stated reason; color: core `palette` with background/secondary_bg/primary/accent/secondary_accent/body_text; typography: CJK + Latin for `heading` and `body` with `css` preview stacks + `body_size` as the body baseline in **px** (every canvas) — **one fixed value per confirmed `delivery_purpose`** (`text` 20 / `balanced` 24 / `presentation` 32), not a range; each typography candidate must include topic-matched `sample_heading` / `sample_heading_latin` / `sample_body` / `sample_body_latin` preview text, never a fixed unrelated industry sample); enumerable `icons` / `formula_policy` (recommended `id`). The still-open page polls, renders Stage 2, and preserves the user's Stage 1 picks. Then attach to the already-running page, do **not** relaunch:+ Page opens at `http://localhost:5050` — the **same port as the Step 6 live preview** (they never run at once: this page shuts down at the end of Step 4). If 5050 is held, the launcher **auto-advances** (5051, …) — read the actual URL from the launch log and report it. The page does **not** close after Stage 1: it shows a "deriving…" state and polls for Stage 2. **Launch or wait failure is non-fatal**: if it fails or times out (flask missing, port blocked, no GUI / remote / web host), do **NOT** troubleshoot — **on any non-zero exit, re-check `result.json` once** for a fresh `status: stage1-confirmed` before dropping to the chat fallback. **On success (exit 0 with a stage-1 result), do not pause or report — go straight to step 3 in the same turn.**+3. **Re-derive Stage 2 from the confirmed anchors, write it, then wait for the design-system handoff — immediately, same turn (the page is polling for it).** Read the stage-1 `result.json` (`status: stage1-confirmed`). Using the user's **actual** confirmed anchors (not your originals), author the design-system candidates and **overwrite** `recommendations.json` with `"stage": "stage2"`: page count (content volume × `delivery_purpose`); color and typography as **generative ≥3-candidate** fields (creative recommendations always offer real choice; fewer than 3 only on the honest-shortfall exception, with a stated reason; color: core `palette` with background/secondary_bg/primary/accent/secondary_accent/body_text; typography: CJK + Latin for `heading` and `body` with `css` preview stacks + `body_size` as the body baseline in **px** (every canvas) — **one fixed value per confirmed `delivery_purpose`** (`text` 20 / `balanced` 24 / `presentation` 32), not a range; each typography candidate must include topic-matched `sample_heading` / `sample_heading_latin` / `sample_body` / `sample_body_latin` preview text, never a fixed unrelated industry sample); enumerable `icons` / `formula_policy` (recommended `id`). The still-open page polls, renders Stage 2, and preserves the user's Stage 1 picks. Then attach to the already-running page, do **not** relaunch:```bash- python3 ${SKILL_DIR}/scripts/confirm_ui/server.py <project_path> --wait-only --wait-stage tier2+ python3 ${SKILL_DIR}/scripts/confirm_ui/server.py <project_path> --wait-only --wait-stage stage2```- This returns when the page writes the tier-2 `result.json` (`status: tier2-confirmed`). On a non-zero exit, re-check `result.json` once before falling back to chat.-4. **Re-derive Stage 3 from the confirmed anchors + design system, then wait for the final confirmation.** Read the stage-2 `result.json`. Author the image and execution recommendations and **overwrite** `recommendations.json` with `"tier": 3`: `image_usage` as one or more source ids (`["ai"]`, `["ai","provided"]`, `["web","placeholder"]`, or `["none"]`; `none` is exclusive); `image_strategy.candidates` as **exactly three non-custom** rendering × palette recommendations from h.5 when `image_usage` includes `ai` (the page adds the fourth Custom card itself); enumerable `image_ai_path` / `generation_mode` and `refine_spec` (recommended `id` / boolean). If the recommendation involves several image sources, keep the source list structured in `recommend.image_usage` and write the usage rationale / page-role guidance into `image_notes` (for example, "封面和章节页用 AI 主视觉,产品页优先用户素材,行业背景页可用网络参考"). Write `image_ai_path` only when `image_usage` includes `ai`. Spot-illustration lean is **not** a candidate field here: it derives from the locked `visual_style`'s illustration propensity and is expressed only in the recommendation rationale / `image_notes`, never as a new confirmation field. Generated-image style palettes are **color behavior only**; final image colors follow the confirmed Stage-2 `color`. Custom image-strategy dimensions are handled by the built-in Custom card, are prose-only, and should not promise a gallery reference image. Then attach to the already-running page, do **not** relaunch (same 600000 ms budget):+ This returns when the page writes the stage-2 `result.json` (`status: stage2-confirmed`). On a non-zero exit, re-check `result.json` once before falling back to chat.+4. **Re-derive Stage 3 from the confirmed anchors + design system, then wait for the final confirmation.** Read the stage-2 `result.json`. Author the image and execution recommendations and **overwrite** `recommendations.json` with `"stage": "stage3"`: `image_usage` as one or more source ids (`["ai"]`, `["ai","provided"]`, `["web","placeholder"]`, or `["none"]`; `none` is exclusive); `image_strategy.candidates` as **exactly three non-custom** rendering × palette recommendations from h.5 when `image_usage` includes `ai` (the page adds the fourth Custom card itself); enumerable `image_ai_path` / `generation_mode` and `refine_spec` (recommended `id` / boolean). If the recommendation involves several image sources, keep the source list structured in `recommend.image_usage` and write the usage rationale / page-role guidance into `image_notes` (for example, "封面和章节页用 AI 主视觉,产品页优先用户素材,行业背景页可用网络参考"). Write `image_ai_path` only when `image_usage` includes `ai`. Spot-illustration lean is **not** a candidate field here: it derives from the locked `visual_style`'s illustration propensity and is expressed only in the recommendation rationale / `image_notes`, never as a new confirmation field. Generated-image style palettes are **color behavior only**; final image colors follow the confirmed Stage-2 `color`. Custom image-strategy dimensions are handled by the built-in Custom card, are prose-only, and should not promise a gallery reference image. Then attach to the already-running page, do **not** relaunch (same 600000 ms budget):```bashpython3 ${SKILL_DIR}/scripts/confirm_ui/server.py <project_path> --wait-only```@@ -415,20 +415,20 @@ When the confirmed `image_usage` does not include `ai` (and no legacy custom proReconcile **without a new blocking wait** — fold the coherent values into `design_spec.md` / `spec_lock.md` and state the adjustment in the §8 next-step handoff (e.g. "you switched to `dark-tech`; the light palette you had left no longer fit, so background / accent were re-derived — tell me if you wanted the original"). Canvas is the explicit exception: font sizes are deliberately **not** rescaled on a canvas change (see strategist §g).-**Opt-out**: if the user has said they don't want the page (e.g. "不要网页" / "just confirm in chat" / "纯聊天确认"), skip the launch entirely (step 2) and present the Eight Confirmations in chat as before — steps 1, 3, 4 still apply (recommendations summary in chat; wait; take chat values).+**Opt-out**: if the user has said they don't want the page (e.g. "不要网页" / "just confirm in chat" / "纯聊天确认"), skip the launch entirely (step 2) and present the Strategist confirmation stage in chat as before — steps 1, 3, 4 still apply (recommendations summary in chat; wait; take chat values).The page is a **confirmation surface only** — Strategist still authors every recommendation; the page never generates content.-**Mandatory — split-mode note** (not a ninth confirmation): after listing the eight confirmation details, you MUST append exactly one short line (rendered in the user's language, prefixed with 💡) about generation mode. Pick the variant by qualitative read of Phase A signals — recommended page count, source-material bulk, whether `topic-research` ran with substantial web-fetch accumulation:+**Mandatory — split-mode note** (not a separate confirmation): after listing the Strategist confirmation stage details, you MUST append exactly one short line (rendered in the user's language, prefixed with 💡) about generation mode. Pick the variant by qualitative read of upstream-load signals — recommended page count, source-material bulk, whether `topic-research` ran with substantial web-fetch accumulation:| Signal read | Line content ||---|---|-| Heavy (long page count / bulky sources / heavy web-fetch accumulation) | State estimated page count and large source size; recommend switching to [split mode](workflows/resume-execute.md) after Step 5 — stop this chat, open a fresh window and input `继续生成 projects/<project_name>` to enter Phase B (SVG generation + export); no response or "continue" = default continuous mode. |+| Heavy (long page count / bulky sources / heavy web-fetch accumulation) | State estimated page count and large source size; recommend switching to [split mode](workflows/resume-execute.md) after Step 5 — stop this chat, open a fresh window and input `继续生成 projects/<project_name>` to enter the execution session (SVG generation + export); no response or "continue" = default continuous mode. || Normal (default) | State scale is moderate, default continuous mode generates in one go; if mid-way window switch is desired, input `继续生成 projects/<project_name>` after Step 5 to switch to [split mode](workflows/resume-execute.md). |This line is required output every run — the user must always see the mode choice exists. Whether to act on it is the user's call. When the Confirm UI is used, this choice also appears as the in-page generation-mode toggle and is captured in `result.json` (`generation_mode`); the chat-summary fallback still prints this line.-**Mandatory — spec-refinement note** (not a ninth confirmation): after the split-mode line, you MUST append one short opt-in line (rendered in the user's language, prefixed with 💡) telling the user they may **refine the spec first** — Strategist will produce the full design spec, then stop for review/revision of any part of it before any generation, via the [refine-spec](workflows/refine-spec.md) workflow. Default is OFF: no request → the spec is written in one go and the pipeline auto-proceeds as usual. Only when the user explicitly asks in chat (e.g. "refine the spec first") or confirms `refine_spec: true` through Confirm UI does the [refine-spec](workflows/refine-spec.md) workflow take over after the Eight Confirmations. This line, like the split-mode line, is required output every run — the user must see the choice exists; whether to act on it is theirs. When the Confirm UI is used, this choice also appears as the in-page refine-spec toggle and is captured in `result.json` (`refine_spec`); the chat-summary fallback still prints this line.+**Mandatory — spec-refinement note** (not a separate confirmation): after the split-mode line, you MUST append one short opt-in line (rendered in the user's language, prefixed with 💡) telling the user they may **refine the spec first** — Strategist will produce the full design spec, then stop for review/revision of any part of it before any generation, via the [refine-spec](workflows/refine-spec.md) workflow. Default is OFF: no request → the spec is written in one go and the pipeline auto-proceeds as usual. Only when the user explicitly asks in chat (e.g. "refine the spec first") or confirms `refine_spec: true` through Confirm UI does the [refine-spec](workflows/refine-spec.md) workflow take over after the Strategist confirmation stage. This line, like the split-mode line, is required output every run — the user must see the choice exists; whether to act on it is theirs. When the Confirm UI is used, this choice also appears as the in-page refine-spec toggle and is captured in `result.json` (`refine_spec`); the chat-summary fallback still prints this line.**Formula rendering policy lives inside item 7 (Typography plan)**:@@ -438,7 +438,7 @@ This line is required output every run — the user must always see the mode cho| `render-all` | Strategist renders every formula-worthy expression as PNG assets || `text-only` | No formula rendering; formulas remain editable text / Unicode |-After the Eight Confirmations are approved and **before outputting `design_spec.md` / `spec_lock.md`**, if the confirmed formula policy is `mixed` or `render-all` and the content contains formula-worthy expressions, Strategist MUST:+After the Strategist confirmation stage is approved and **before outputting `design_spec.md` / `spec_lock.md`**, if the confirmed formula policy is `mixed` or `render-all` and the content contains formula-worthy expressions, Strategist MUST:1. Identify explicit LaTeX and any source expressions that should be faithfully structured as formulas.2. Write `<project_path>/images/formula_manifest.json` with only the formulas selected for rendering.@@ -466,9 +466,9 @@ python3 ${SKILL_DIR}/scripts/analyze_images.py <project_path>/images**✅ Checkpoint — Phase deliverables complete, auto-proceed to next step**:```markdown## ✅ Strategist Phase Complete-- [x] Read the auto-extracted facts already in `analysis/` (e.g. `source_profile.json`) before the Eight Confirmations-- [x] Eight Confirmations completed (user confirmed via Confirm UI `result.json` or chat fallback)-- [x] Split-mode note appended below the eight items (heavy or normal variant)+- [x] Read the auto-extracted facts already in `analysis/` (e.g. `source_profile.json`) before the Strategist confirmation stage+- [x] Strategist confirmation stage completed (user confirmed via Confirm UI `result.json` or chat fallback)+- [x] Split-mode note appended below the confirmation fields (heavy or normal variant)- [x] Spec-refinement opt-in line appended (default OFF; only the user's explicit request enters the refine-spec workflow)- [x] Design Specification & Content Outline generated- [x] Execution lock (spec_lock.md) generated@@ -529,13 +529,13 @@ Workflow:- [x] analyze_images.py re-run so image_analysis.csv covers the acquired web / AI / sliced images```-**Default — auto-proceed to Step 6.** Only when the user's Step 4 response explicitly opted into split mode (in chat or via Confirm UI `result.json` with `generation_mode: "split"`), output the Phase A hand-off below and stop this conversation:+**Default — auto-proceed to Step 6.** Only when the user's Step 4 response explicitly opted into split mode (in chat or via Confirm UI `result.json` with `generation_mode: "split"`), output the planning-session handoff below and stop this conversation:```markdown- ## ✅ Phase A Complete+ ## ✅ Planning Session Complete- [x] Spec: `design_spec.md`, `spec_lock.md`- [x] Resources: `sources/`, `images/`, `templates/`- - [ ] **Next**: open a fresh chat window and input `继续生成 projects/<project_name>` to enter Phase B via the [`resume-execute`](workflows/resume-execute.md) workflow.+ - [ ] **Next**: open a fresh chat window and input `继续生成 projects/<project_name>` to enter the execution session via the [`resume-execute`](workflows/resume-execute.md) workflow.```> On acquisition failure, do NOT halt — follow the Failure Handling rule in [image-base.md](references/image-base.md) §5: retry once, then mark the row `Needs-Manual`, report to user, and continue to the checkpoint above.@@ -612,7 +612,7 @@ python3 ${SKILL_DIR}/scripts/svg_quality_checker.py <project_path>🚧 **Image readiness GATE** (when Step 5 left ai rows in `Needs-Manual`): every expected file must exist at `project/images/<filename>` before running 7.1.-**Failure recovery**: if a Step 7 command fails, fix the owning source artifact and resume from the failed sub-step per [`workflows/failure-recovery.md`](workflows/failure-recovery.md). Do not restart Phase A unless the owning source changed.+**Failure recovery**: if a Step 7 command fails, fix the owning source artifact and resume from the failed sub-step per [`workflows/failure-recovery.md`](workflows/failure-recovery.md). Do not restart the planning session unless the owning source changed.> If files are missing: PAUSE, list the missing filenames, point the user to `images/image_prompts.md` (each `### Image N:` block is paste-ready for ChatGPT / Gemini / Midjourney; auto-generated from `image_prompts.json`) and the required placement `project/images/<filename>`. Resume Step 7.1 only after all expected files are in place. `finalize_svg.py` and `svg_to_pptx.py` do not detect missing files at this layer — proceeding with gaps produces a deck with broken image references.diff --git a/skills/ppt-master/references/artifact-ownership.md b/skills/ppt-master/references/artifact-ownership.mdindex b0b8009f..f3d7e84e 100644--- a/skills/ppt-master/references/artifact-ownership.md+++ b/skills/ppt-master/references/artifact-ownership.md@@ -22,7 +22,7 @@ Global artifact ownership rules for PPT Master projects.| `images/` | Runtime image pool | User, extracted, AI, web, formula, slice, EMF/WMF assets | Step 5 writes here; `analysis/image_analysis.csv` derives from current contents || `icons/` | Project icon inventory | Icons copied by `icon_sync.py` for this project | Executor uses locked project icons; exporter may fall back to global library only as documented || `templates/` | Project template reference | Step 3 imported specs, template SVGs, and non-image assets | Strategist/Executor read only when Step 3 is triggered |-| `confirm_ui/recommendations.json` | Confirmation proposal | Strategist-authored confirmation payload | Confirm UI reads; rewritten between Tier 1 and Tier 2 |+| `confirm_ui/recommendations.json` | Confirmation proposal | Strategist-authored confirmation payload | Confirm UI reads; rewritten between Stage 1, Stage 2, and Stage 3 || `confirm_ui/result.json` | Confirmation result | User-confirmed values | Strategist treats final result as authoritative over recommendations || `svg_output/` | Author source | Main-agent handwritten SVG pages | Quality checker and native PPTX export read this as canonical page source || `notes/total.md` | Speaker-note source | Complete notes before splitting | Step 6 writes; Step 7.1 splits |diff --git a/skills/ppt-master/references/image-layout-spec.md b/skills/ppt-master/references/image-layout-spec.mdindex 6626d629..723135b0 100644--- a/skills/ppt-master/references/image-layout-spec.md+++ b/skills/ppt-master/references/image-layout-spec.md@@ -22,7 +22,7 @@ Layout rules for pages where the image is placed **side-by-side with body text**7. Fill results into the Design Specification's image resource list```-**When to run**: if image approach includes "B) User-provided", run the scan and populate the image resource list after the Strategist's Eight Confirmations and before content analysis / outlining.+**When to run**: if image approach includes "B) User-provided", run the scan and populate the image resource list after the Strategist's confirmation stage and before content analysis / outlining.---diff --git a/skills/ppt-master/references/strategist.md b/skills/ppt-master/references/strategist.mdindex d1074ad7..11794c06 100644--- a/skills/ppt-master/references/strategist.md+++ b/skills/ppt-master/references/strategist.md@@ -8,7 +8,7 @@ As a top-tier AI presentation strategist, receive source documents, perform cont| Previous Step | Current | Next Step ||--------------|---------|-----------|-| Project creation + Template option confirmed | **Strategist**: Eight Confirmations + Design Spec | Image_Generator or Executor |+| Project creation + Template option confirmed | **Strategist**: Strategist confirmation stage + Design Spec | Image_Generator or Executor |---@@ -18,13 +18,13 @@ As a top-tier AI presentation strategist, receive source documents, perform cont----## 1. Eight Confirmations Process+## 1. Strategist Confirmation Stage🚧 **GATE — Mandatory read first**: `read_file templates/design_spec_reference.md` before any analysis or writing. The design_spec.md output MUST follow that template's 11-section structure exactly. After writing, self-check each section is present: I Project Info → II Canvas → III Visual Theme → IV Typography → V Layout → VI Icon → VII Visualization → VIII Image → IX Outline → X Speaker Notes → XI Tech Constraints.-⛔ **BLOCKING**: After the read, present professional recommendations for the eight items below and wait for explicit user confirmation.+⛔ **BLOCKING**: After the read, present professional recommendations for the confirmation fields below and wait for explicit user confirmation.-**Three-stage confirmation (the default Confirm UI flow; chat mirrors it).** The eight items split into a dependency order, confirmed in three passes:+**Three-stage confirmation (the default Confirm UI flow; chat mirrors it).** The confirmation fields split into a dependency order, confirmed in three passes:| Stage | Items | Role ||---|---|---|@@ -86,7 +86,7 @@ Write the locked value to `spec_lock.md` `- mode:` and record the rationale in `🚧 **GATE**: read [`visual-styles/_index.md`](./visual-styles/_index.md) before recommending.-The deck's **visual aesthetic** — shape language, decoration density, whitespace rhythm, typographic character, texture. Anchors the downstream confirmations e (Color), f (Icon), g (Typography), h (Image). Lock one preset from the catalog, or `custom`.+The 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: map to the closest preset (or `custom` with a `visual_style_behavior` paragraph) and lock directly. **Skip the spectrum below** — do not re-offer choice they already made.@@ -144,7 +144,7 @@ See [`../templates/icons/README.md`](../templates/icons/README.md) for the curre> **Mandatory rules when choosing C**:>-> **At the eight-confirmation stage — decide the library only. Do NOT run `ls | grep` yet.**+> **At the Strategist confirmation stage — decide the library only. Do NOT run `ls | grep` yet.**>> 1. **Pick exactly one stylistic library** — read the source material, then choose the library whose visual character best serves the deck:> - **`chunk-filled`** — fill, straight-line geometry (M/L/H/V/Z only); sharp right angles; heavy, solid, architectural@@ -155,7 +155,7 @@ See [`../templates/icons/README.md`](../templates/icons/README.md) for the curre> - **Brand-logo exception**: `simple-icons` is NOT a stylistic library. Add it to the deck's icon inventory **only when** the deck genuinely contains real company / product / service brand marks (customer logos, tech-stack icons, social handles). Never substitute it for a missing generic icon.> 2. **Stroke weight lock (stroke-style libraries only)** — for stroke-based libraries (currently `tabler-outline`), pick one deck-wide value from `{1.5, 2, 3}` (default `2`). For heavier presence, switch library instead of going above `3`.>-> **After all eight confirmations are approved — when writing `design_spec.md` §VI / `spec_lock.md`**, then materialize the icon inventory:+> **After the Strategist confirmation stage is approved — when writing `design_spec.md` §VI / `spec_lock.md`**, then materialize the icon inventory:>> 3. Enumerate the concepts the deck actually needs (home, chart, users, …) based on the confirmed outline.> 4. Search for each concept's filename in the chosen library: `ls skills/ppt-master/templates/icons/<chosen-library>/ | grep <keyword>`@@ -304,7 +304,7 @@ Formula rendering is part of Typography confirmation. Recommend one policy and l**Forbidden — invented math**: formula assets must faithfully structure source content. Do not create a new equation just to make a slide look more academic.-**Manifest step**: After the Eight Confirmations and before writing `design_spec.md`, if policy is `mixed` or `render-all` and formulas are selected:+**Manifest step**: After the Strategist confirmation stage and before writing `design_spec.md`, if policy is `mixed` or `render-all` and formulas are selected:```bashmkdir -p <project_path>/images@@ -673,7 +673,7 @@ The catalog covers **both data charts and structural information designs**. A "mThe most common Strategist failure mode is missing the structural half — treating "chart" as "numeric chart only" and leaving team / agenda / principles / journey pages as text-only when a template would fit. Read the catalog with both lenses.> **Reading is mandatory; the catalog is a starting point, not a copy target.**-> - Fully read `templates/charts/charts_index.json` **before drafting the Eight Confirmations** — the read happens up front, not when you sit down to write Section VII. The file contains `meta` + `charts.<key>.summary` only; each `summary` is a selection rule (`"Pick for … Skip if …"`), not a description. There is **no category, quickLookup, or keyword index** — selection is done by semantically matching each page's content shape against all 76 summaries in one pass.+> - Fully read `templates/charts/charts_index.json` **before drafting the Strategist confirmation stage** — the read happens up front, not when you sit down to write Section VII. The file contains `meta` + `charts.<key>.summary` only; each `summary` is a selection rule (`"Pick for … Skip if …"`), not a description. There is **no category, quickLookup, or keyword index** — selection is done by semantically matching each page's content shape against all 76 summaries in one pass.> - Not every page needs a chart. When a page's information structure matches a catalog entry, **use that template as a structural starting point** — keep the visualization type and core layout logic, then adapt composition, density, color, decoration, and accompanying elements to fit this deck's content and visual tone. Free adjustment is encouraged; what is forbidden is (a) generating without reading the catalog, and (b) blind verbatim mimicry that ignores the page's actual content weight.>> **Workflow**:diff --git a/skills/ppt-master/references/svg-image-embedding.md b/skills/ppt-master/references/svg-image-embedding.mdindex 31a32424..fe5b5917 100644--- a/skills/ppt-master/references/svg-image-embedding.md+++ b/skills/ppt-master/references/svg-image-embedding.md@@ -8,7 +8,7 @@ Technical spec and workflow for adding images to SVG files.## Image Resource List Format-Defined in the Design Specification & Content Outline; each image carries an `Acquire Via` field plus a status annotation. This file is authoritative for status names and SVG embedding behavior. If image approach includes "B) User-provided": run `analyze_images.py` right after the Eight Confirmations and complete the list before outputting the design spec.+Defined in the Design Specification & Content Outline; each image carries an `Acquire Via` field plus a status annotation. This file is authoritative for status names and SVG embedding behavior. If image approach includes "B) User-provided": run `analyze_images.py` right after the Strategist confirmation stage and complete the list before outputting the design spec.```markdown| Filename | Dimensions | Purpose | Type | Acquire Via | Status | Reference |diff --git a/skills/ppt-master/scripts/confirm_ui/__init__.py b/skills/ppt-master/scripts/confirm_ui/__init__.pyindex 52480fc0..79e12956 100644--- a/skills/ppt-master/scripts/confirm_ui/__init__.py+++ b/skills/ppt-master/scripts/confirm_ui/__init__.py@@ -1 +1 @@-"""PPT Master - interactive Eight Confirmations UI (Step 4)."""+"""PPT Master - interactive Strategist confirmation stage UI (Step 4)."""diff --git a/skills/ppt-master/scripts/confirm_ui/server.py b/skills/ppt-master/scripts/confirm_ui/server.pyindex 6c10efdd..1b4a4efd 100644--- a/skills/ppt-master/scripts/confirm_ui/server.py+++ b/skills/ppt-master/scripts/confirm_ui/server.py@@ -1,8 +1,8 @@#!/usr/bin/env python3"""-PPT Master - Eight Confirmations UI Server (Step 4)+PPT Master - Strategist confirmation stage UI Server (Step 4)-Lightweight Flask backend for the interactive, visual Eight Confirmations page.+Lightweight Flask backend for the interactive, visual Strategist confirmation stage page.Strategist writes its recommendations to``<project>/confirm_ui/recommendations.json``; this server renders them as aclickable page (color swatches, live font previews, candidate picks). On@@ -11,7 +11,7 @@ submit it writes the user's final choices toThis is the confirmation surface only. The chat fallback always remains valid:if the browser cannot open (remote / headless / web host), the AI presents the-same Eight Confirmations in chat.+same Strategist confirmation stage in chat.See scripts/docs/confirm_ui.md for the round-trip data contract and schema.@@ -137,12 +137,42 @@ def _wait_for_result(def _result_stage(result_file: Path) -> Optional[str]:- """Return the ``stage`` field of result.json, or None."""+ """Return the canonical ``stage`` field of result.json, or None."""try:data = json.loads(result_file.read_text(encoding='utf-8'))except (OSError, json.JSONDecodeError):return None- return data.get('stage') if isinstance(data, dict) else None+ return _stage_key(data.get('stage')) if isinstance(data, dict) else None+++def _stage_key(value: object) -> Optional[str]:+ """Normalize current stage names while accepting legacy tier values."""+ if value is None:+ return None+ raw = str(value).strip().lower()+ if raw in {'1', 'stage1', 'tier1'}:+ return 'stage1'+ if raw in {'2', 'stage2', 'tier2'}:+ return 'stage2'+ if raw in {'3', 'stage3', 'tier3'}:+ return 'stage3'+ if raw == 'final':+ return 'final'+ return None+++def _recommendation_stage(data: dict) -> int:+ """Return recommendations.json stage number, with legacy tier fallback."""+ stage = _stage_key(data.get('stage'))+ if not stage and 'tier' in data:+ stage = _stage_key(data.get('tier'))+ if stage == 'stage1':+ return 1+ if stage == 'stage2':+ return 2+ if stage == 'stage3':+ return 3+ return 0# Stage-1 anchors and Stage-2 design-system choices. On later pages these sections@@ -172,11 +202,7 @@ def _merge_confirmed_choices(data: dict, result_file: Path) -> None:for key in _ANCHOR_VALUE_KEYS:if key in res:data[key] = {'value': res.get(key) or ''}- try:- tier = int(data.get('tier') or 0)- except (TypeError, ValueError):- tier = 0- if tier < 3:+ if _recommendation_stage(data) < 3:returnfor key in _DESIGN_RECOMMEND_KEYS:if res.get(key) not in (None, ''):@@ -472,11 +498,7 @@ def create_app(# Later stages render only downstream sections, so fold earlier confirmed# choices from result.json back in. A refresh / reopen then re-inits from# the user's choices instead of catalog defaults.- try:- tier = int(data.get('tier') or 0)- except (TypeError, ValueError):- tier = 0- if tier >= 2 and result_file.exists():+ if _recommendation_stage(data) >= 2 and result_file.exists():_merge_confirmed_choices(data, result_file)# The page polls this endpoint after a stage-1 confirm until the AI# overwrites the file with the re-derived stage-2 recommendations, so it@@ -496,15 +518,20 @@ def create_app(# Staged flow: stage-1 / stage-2 submits record intermediate choices but do# NOT close the page. Only a final submit is a full confirmation. A# payload with no stage is a single-pass confirmation (chat-opt-out parity).- stage = result.get('stage')- result['status'] = f'{stage}-confirmed' if stage in {'tier1', 'tier2'} else 'confirmed'+ stage = _stage_key(result.get('stage'))+ if stage in {'stage1', 'stage2'}:+ result['stage'] = stage+ result['status'] = f'{stage}-confirmed'+ else:+ result['stage'] = 'final'+ result['status'] = 'confirmed'result['confirmed_at'] = time.strftime('%Y-%m-%dT%H:%M:%S')result_file = confirm_dir / RESULT_NAMEresult_file.write_text(json.dumps(result, ensure_ascii=False, indent=2),encoding='utf-8',)- logger.info('%s confirmation written to %s', stage or 'final', result_file)+ logger.info('%s confirmation written to %s', result['stage'], result_file)return jsonify({'status': 'ok'})return app@@ -512,7 +539,7 @@ def create_app(def build_parser() -> argparse.ArgumentParser:parser = argparse.ArgumentParser(- description='PPT Master Eight Confirmations UI',+ description='PPT Master Strategist confirmation stage UI',formatter_class=argparse.RawDescriptionHelpFormatter,)parser.add_argument('project_dir', help='Path to project directory')@@ -535,9 +562,9 @@ def build_parser() -> argparse.ArgumentParser:'this project and wait for an already-open page to write result.json.',)parser.add_argument(- '--wait-stage', choices=('tier2', 'final'), default='final',+ '--wait-stage', default='final', metavar='{stage2,final}',help='With --wait-only, wait for this result.json stage (default: final). '- 'Use tier2 for the middle handoff in the three-stage flow.',+ 'Use stage2 for the middle handoff in the three-stage flow.',)parser.add_argument('--wait-timeout', type=int, default=WAIT_TIMEOUT_DEFAULT,@@ -571,20 +598,24 @@ def main(argv: Optional[list[str]] = None) -> int:if not project_path.is_dir():logger.error('%s is not a directory', project_path)return 1+ wait_stage = _stage_key(args.wait_stage)+ if wait_stage not in {'stage2', 'final'}:+ logger.error('--wait-stage must be stage2 or final')+ return 2# Step 4 cleanup: stop any lingering confirm server and exit. Independent of# recommendations.json (the page may never have been confirmed).if args.shutdown:return _shutdown_existing(project_path / LOCK_FILE_NAME)- # Stage-2 wait: attach to the server already serving this project (launched by- # the first --wait) and block until the page writes the final result.json.+ # Staged wait: attach to the server launched by the first --wait and block+ # until the page writes the requested intermediate or final result.json.if args.wait_only:return _wait_only_for_result(project_path / CONFIRM_DIR_NAME / RESULT_NAME,project_path / LOCK_FILE_NAME,args.wait_timeout,- args.wait_stage,+ wait_stage,)rec_file = project_path / CONFIRM_DIR_NAME / RECOMMENDATIONS_NAMEdiff --git a/skills/ppt-master/scripts/confirm_ui/static/app.js b/skills/ppt-master/scripts/confirm_ui/static/app.jsindex 1ad01fa4..79c000d0 100644--- a/skills/ppt-master/scripts/confirm_ui/static/app.js+++ b/skills/ppt-master/scripts/confirm_ui/static/app.js@@ -1,4 +1,4 @@-/* PPT Master - Eight Confirmations UI+/* PPT Master - Strategist confirmation stage UI* Finite/enumerable fields (canvas, mode, visual style, icons, image usage,* AI source, formula policy, generation mode) list ALL options from* /static/catalogs.json with the AI's recommendation marked. Open/generative@@ -16,7 +16,6 @@stage_anchors: "Stage 1 · Direction",stage_design: "Stage 2 · Design system",stage_images: "Stage 3 · Images & execution",- stage_realization: "Stage 2 · Design system",loading: "Loading…",load_error: "Could not load recommendations.json. The AI must write it before launch.",btn_confirm: "Confirm",@@ -122,7 +121,6 @@stage_anchors: "ステージ 1 · 方向性",stage_design: "ステージ 2 · デザインシステム",stage_images: "ステージ 3 · 画像と実行方法",- stage_realization: "ステージ 2 · デザインシステム",loading: "読み込み中…",load_error: "recommendations.json を読み込めませんでした。起動前にAIが書き込む必要があります。",btn_confirm: "確定",@@ -228,7 +226,6 @@stage_anchors: "第一阶段 · 方向确认",stage_design: "第二阶段 · 设计系统",stage_images: "第三阶段 · 图片与执行方式",- stage_realization: "第二阶段 · 设计系统",loading: "加载中…",load_error: "无法加载推荐文件,需在启动前写入。",btn_confirm: "确认",@@ -644,7 +641,7 @@}// Section numbers run 1..N within the stage currently rendered; the counter is- // reset at the top of renderForTier. The legacy `num` arg is ignored so each+ // reset at the top of renderForStage. The legacy `num` arg is ignored so each// stage numbers its own sections cleanly (stage 2 is not a continuation of 1).var _secCounter = 0;function section(num, titleKey, noteText) {@@ -1513,7 +1510,7 @@}}- // Combined color + typography + icon preview — not a ninth confirmation, just a+ // Combined color + typography + icon preview — not a separate confirmation, just a// live "overall impression" of the style choices made above. Kept// deliberately abstract (a style chip, not a slide layout); page layout// preview is the live-preview server's job (Step 6).@@ -1888,22 +1885,31 @@// Stage of the staged confirm flow:// 1 = direction anchors, 2 = design system, 3 = images/execution,- // "all" = legacy single-pass (recommendations.json carried no `tier`).+ // "all" = legacy single-pass (recommendations.json carried no stage).var STAGE = 1;- function stageTitle(tier) {- if (tier === 1) return t("stage_anchors");- if (tier === 2) return t("stage_design");- if (tier === 3) return t("stage_images");+ function stageNumber(data) {+ var raw = data && data.stage != null ? data.stage : (data && data.tier);+ raw = String(raw == null ? "" : raw).toLowerCase();+ if (raw === "1" || raw === "stage1" || raw === "tier1") return 1;+ if (raw === "2" || raw === "stage2" || raw === "tier2") return 2;+ if (raw === "3" || raw === "stage3" || raw === "tier3") return 3;+ return "all";+ }++ function stageTitle(stage) {+ if (stage === 1) return t("stage_anchors");+ if (stage === 2) return t("stage_design");+ if (stage === 3) return t("stage_images");return t("page_title");}- function renderForTier(tier) {+ function renderForStage(stage) {var host = document.getElementById("sections");host.innerHTML = "";_secCounter = 0;var heading = document.querySelector("#topbar .topbar-titles h1");- if (heading) heading.textContent = stageTitle(tier);+ if (heading) heading.textContent = stageTitle(stage);// Detach the previous preview's repaint closures before the sections// re-render: color/typography auto-select would otherwise call them and// write to now-detached nodes until renderStylePreview remounts them.@@ -1914,14 +1920,14 @@refreshSizeInputs = function () {};var previewHost = document.getElementById("topbar-preview");if (previewHost) previewHost.innerHTML = "";- if (tier === 1) {+ if (stage === 1) {if (previewHost) renderDirectionPreview(previewHost);// Direction anchors — Stage 2 is re-derived from these.// Delivery purpose rides inside renderAudience (§c key info).renderCanvas(host);renderAudience(host);renderStyle(host);- } else if (tier === 2) {+ } else if (stage === 2) {if (previewHost) renderStylePreview(previewHost);renderPages(host);// Group the three sections reflected by the fixed preview strip.@@ -1930,7 +1936,7 @@renderIcons(styleGroup);renderTypography(styleGroup);host.appendChild(styleGroup);- } else if (tier === 3) {+ } else if (stage === 3) {if (previewHost) renderImageStrategyPreview(previewHost);renderImages(host);renderMode(host);@@ -1951,16 +1957,16 @@renderMode(host);renderRefine(host);}- updateActionBar(tier);+ updateActionBar(stage);}- function renderAll() { renderForTier(STAGE); }+ function renderAll() { renderForStage(STAGE); }- function updateActionBar(tier) {+ function updateActionBar(stage) {var btn = document.getElementById("btn-confirm");btn.disabled = false;// Stage 1/2 advance; Stage 3 / single-pass confirm.- btn.textContent = (tier === 1 || tier === 2) ? t("btn_next") : t("btn_confirm");+ btn.textContent = (stage === 1 || stage === 2) ? t("btn_next") : t("btn_confirm");}// ---- state init (once) ----------------------------------------------@@ -1973,7 +1979,7 @@return recOrFirst(field, catList);}- function initTier1State() {+ function initStage1State() {STATE.canvas = pick("canvas", CAT.canvas);STATE.audience = (REC.audience && REC.audience.value) || "";STATE.content_divergence = (REC.content_divergence && REC.content_divergence.value) || ""; // free text; blank = balanced default@@ -1985,10 +1991,10 @@}// Stage-2 fields are (re-)read from the recommendations. At boot they come from- // whatever recommendations.json carried; after a stage-1 confirm enterTier()+ // whatever recommendations.json carried; after a stage-1 confirm enterStage()// calls this again with the re-derived candidates. Stage-1 STATE is preserved// across the single-session transition — this never resets the anchors.- function initTier2State() {+ function initStage2State() {STATE.page_count = (REC.page_count && REC.page_count.value != null) ? String(REC.page_count.value) : (STATE.page_count || "");var cc = (REC.color && REC.color.candidates) || [];@@ -2018,7 +2024,7 @@}}- function initTier3State() {+ function initStage3State() {var rawImageUsage = recValue("image_usage");STATE.image_usage = selectedImageUsageIds(rawImageUsage);if (!STATE.image_usage.length) {@@ -2032,9 +2038,9 @@}function initState() {- initTier1State();- initTier2State();- initTier3State();+ initStage1State();+ initStage2State();+ initStage3State();}// ---- confirm + close -------------------------------------------------@@ -2046,9 +2052,9 @@}// ---- staged submit + re-derive transitions --------------------------- function tier1Payload() {+ function stage1Payload() {var payload = {- stage: "tier1",+ stage: "stage1",canvas: STATE.canvas,audience: STATE.audience,content_divergence: STATE.content_divergence,@@ -2060,9 +2066,9 @@return payload;}- function tier2Payload() {+ function stage2Payload() {var payload = {- stage: "tier2",+ stage: "stage2",canvas: STATE.canvas,audience: STATE.audience,content_divergence: STATE.content_divergence,@@ -2079,7 +2085,7 @@return payload;}- function submitStage(payload, nextTier) {+ function submitStage(payload, nextStage) {var btn = document.getElementById("btn-confirm");btn.disabled = true;fetch("/api/confirm", {@@ -2089,16 +2095,16 @@}).then(function (r) {if (!r.ok) throw new Error("stage submit failed");showDeriving();- pollForTier(nextTier);+ pollForStage(nextStage);}).catch(function () {btn.disabled = false;document.getElementById("confirm-status").textContent = t("error_retry");});}- function submitTier1() { submitStage(tier1Payload(), 2); }+ function submitStage1() { submitStage(stage1Payload(), 2); }- function submitTier2() { submitStage(tier2Payload(), 3); }+ function submitStage2() { submitStage(stage2Payload(), 3); }function showDeriving() {document.getElementById("sections").style.display = "none";@@ -2109,30 +2115,30 @@}// Poll the recommendations endpoint (no-store) until the AI overwrites it with- // the next re-derived tier, then render it in the same session.- function pollForTier(nextTier) {+ // the next re-derived stage, then render it in the same session.+ function pollForStage(nextStage) {fetch("/api/recommendations", { cache: "no-store" }).then(function (r) { if (!r.ok) throw new Error("poll failed"); return r.json(); }).then(function (data) {- if (data && String(data.tier) === String(nextTier)) { enterTier(data, nextTier); }- else { setTimeout(function () { pollForTier(nextTier); }, 1200); }+ if (data && stageNumber(data) === nextStage) { enterStage(data, nextStage); }+ else { setTimeout(function () { pollForStage(nextStage); }, 1200); }}).catch(function (err) {var l = document.getElementById("loading");if (l) l.textContent = t("load_error") + " " + (err && err.message ? err.message : "");- setTimeout(function () { pollForTier(nextTier); }, 1500);+ setTimeout(function () { pollForStage(nextStage); }, 1500);});}- function enterTier(data, tier) {+ function enterStage(data, stage) {REC = data;- if (tier === 2) initTier2State();- if (tier === 3) initTier3State();- STAGE = tier;+ if (stage === 2) initStage2State();+ if (stage === 3) initStage3State();+ STAGE = stage;document.getElementById("loading").style.display = "none";document.getElementById("sections").style.display = "block";document.getElementById("actionbar").style.display = "flex";document.getElementById("confirm-status").textContent = "";- renderForTier(tier);+ renderForStage(stage);}function confirm() {@@ -2277,8 +2283,8 @@if (!langMenu.hidden) setMenuOpen(false);});document.getElementById("btn-confirm").addEventListener("click", function () {- if (STAGE === 1) submitTier1();- else if (STAGE === 2) submitTier2();+ if (STAGE === 1) submitStage1();+ else if (STAGE === 2) submitStage2();else confirm();});@@ -2298,9 +2304,8 @@if (!hasStored) { LANG = REC.lang; applyStaticTranslations(); refreshLangToggle(toggleBtn); }}initState();- // tier 1 / 2 / 3 from the recommendations; absent → legacy single-pass.- STAGE = (String(REC.tier) === "1") ? 1 :- (String(REC.tier) === "2" ? 2 : (String(REC.tier) === "3" ? 3 : "all"));+ // stage 1 / 2 / 3 from the recommendations; absent → legacy single-pass.+ STAGE = stageNumber(REC);document.getElementById("loading").style.display = "none";document.getElementById("sections").style.display = "block";document.getElementById("actionbar").style.display = "flex";diff --git a/skills/ppt-master/scripts/confirm_ui/static/catalogs.json b/skills/ppt-master/scripts/confirm_ui/static/catalogs.jsonindex 4fb5b48d..b62ed9d6 100644--- a/skills/ppt-master/scripts/confirm_ui/static/catalogs.json+++ b/skills/ppt-master/scripts/confirm_ui/static/catalogs.json@@ -573,7 +573,7 @@"label_en": "balanced / business","label_ja": "バランス型・ビジネス","desc_zh": "既投影也阅读(路演、业务评审)。默认档,正文 24px。",- "desc_en": "Both projected and read (roadshow, business review). The default tier, body 24px.",+ "desc_en": "Both projected and read (roadshow, business review). The default option, body 24px.","desc_ja": "投影でも閲覧でも使う用途(ロードショー、事業レビュー)。デフォルトの選択、本文24px。"},{diff --git a/skills/ppt-master/scripts/docs/confirm_ui.md b/skills/ppt-master/scripts/docs/confirm_ui.mdindex 48b4010c..17d0d06b 100644--- a/skills/ppt-master/scripts/docs/confirm_ui.md+++ b/skills/ppt-master/scripts/docs/confirm_ui.md@@ -1,6 +1,6 @@-# Confirm UI — Eight Confirmations Page+# Confirm UI — Strategist Confirmation Stage Page-> The interactive, visual surface for SKILL.md Step 4 (the Eight Confirmations). Enumerable fields list **all** options from a catalog with the AI's recommendation badged; generative fields (color, typography, generated-image style) show **≥3** AI candidates (creative recommendations always offer real choice — same rule as the h.5 image strategy; fewer only on the honest-shortfall exception, with a stated reason). Fields whose universe is open (canvas, mode, visual style, icons) also get a **Custom** box; image usage is a multi-select source list plus a free-text `image_notes` box. Fully closed fields (AI source when applicable, formula policy, generation mode, refine spec) do not. The AI writes its recommendation to `recommendations.json`; the user's final choices are written back to `result.json` for the AI to read. On confirm the page saves the result and shuts the server down (auto-close). The chat path is always a valid fallback — if the browser cannot open (remote / headless / web host), the AI presents the same confirmations in chat.+> The interactive, visual surface for SKILL.md Step 4 (the Strategist confirmation stage). Enumerable fields list **all** options from a catalog with the AI's recommendation badged; generative fields (color, typography, generated-image style) show **≥3** AI candidates (creative recommendations always offer real choice — same rule as the h.5 image strategy; fewer only on the honest-shortfall exception, with a stated reason). Fields whose universe is open (canvas, mode, visual style, icons) also get a **Custom** box; image usage is a multi-select source list plus a free-text `image_notes` box. Fully closed fields (AI source when applicable, formula policy, generation mode, refine spec) do not. The AI writes its recommendation to `recommendations.json`; the user's final choices are written back to `result.json` for the AI to read. On confirm the page saves the result and shuts the server down (auto-close). The chat path is always a valid fallback — if the browser cannot open (remote / headless / web host), the AI presents the same staged confirmation in chat.## Authority and Scope@@ -22,7 +22,7 @@```bashpython3 scripts/confirm_ui/server.py <project_path> --daemon --wait # launch + wait for Stage 1-python3 scripts/confirm_ui/server.py <project_path> --wait-only --wait-stage tier2 # Stage 2: wait for the design-system handoff+python3 scripts/confirm_ui/server.py <project_path> --wait-only --wait-stage stage2 # Stage 2: wait for the design-system handoffpython3 scripts/confirm_ui/server.py <project_path> --wait-only # Stage 3: wait for the final resultpython3 scripts/confirm_ui/server.py <project_path> --daemonpython3 scripts/confirm_ui/server.py <project_path> --daemon --port 5051@@ -34,7 +34,7 @@ python3 scripts/confirm_ui/server.py <project_path> --shutdown # Step 4 clean- Binds `127.0.0.1:5050` by default — or the next free port if another project already holds it (the launch log prints the actual URL) — and auto-opens the browser (suppress with `--no-browser`). `--port <other>` forces a specific port.- **Shares port 5050 with the live preview server** (`svg_editor/server.py`). The two never run at once: confirm is Step 4, live preview is Step 6, and Step 4 always shuts this server down on exit (see `--shutdown`) so the port is free. One port = one forward rule for the whole pipeline. They still keep **separate processes and locks** (`.confirm_ui.lock` vs `.live_preview.lock`).- `--daemon` starts the Flask process in the background; add `--wait` in the main pipeline so the parent command returns only after the page writes a fresh `result.json`. The `--wait` budget defaults to **590 s** (`--wait-timeout`), kept under the typical 600 s tool ceiling — run the launch with a long tool timeout (≈600000 ms). On timeout the parent returns non-zero but the detached server keeps running, so the caller must re-check `result.json` once before the chat fallback (a slow user may confirm just after the wait returns).-- `--wait-only` does **not** launch a server — it attaches to the page already running from the first `--daemon --wait` and blocks until the page writes the requested stage. Use `--wait-stage tier2` for the middle design-system handoff, then the default `--wait-stage final` for the final Stage-3 confirmation. It keys on the **stage alone** (no mtime gate), because a user may submit before this wait command is issued. Same `--wait-timeout` budget; on timeout / server-gone it returns non-zero and the caller re-checks `result.json` once before the chat fallback.+- `--wait-only` does **not** launch a server — it attaches to the page already running from the first `--daemon --wait` and blocks until the page writes the requested stage. Use `--wait-stage stage2` for the middle design-system handoff, then the default `--wait-stage final` for the final Stage-3 confirmation. It keys on the **stage alone** (no mtime gate), because a user may submit before this wait command is issued. Same `--wait-timeout` budget; on timeout / server-gone it returns non-zero and the caller re-checks `result.json` once before the chat fallback.- `--shutdown` stops a confirm server left running for this project and exits — **idempotent** (a no-op when nothing is running). Tries a graceful `/api/shutdown`, falls back to killing the recorded pid, then clears the lock. SKILL.md Step 4 runs this on every path (page-confirm or chat-fallback) so the page never lingers on the shared port before live preview starts.- Refuses to start unless `<project_path>/confirm_ui/recommendations.json` exists (except `--shutdown`, which needs no recommendations).- Per-project lock at `<project_path>/.confirm_ui.lock` — duplicate launches are refused; stale locks (dead pid) are overwritten.@@ -68,21 +68,22 @@ Both files live under `<project_path>/confirm_ui/`.### Three-stage flow-The page runs as a **three-stage wizard in one browser session**. `recommendations.json` carries a top-level `"tier"` as the internal stage selector retained for backward compatibility:+The page runs as a **three-stage wizard in one browser session**. `recommendations.json` carries a top-level `"stage"` selector. Legacy payloads that still carry `"tier"` are accepted as read-only compatibility input, but new files must use `stage`.-| Internal `tier` | Page renders | Button | On submit |+| `recommendations.json stage` | Page renders | Button | On submit ||---|---|---|---|-| `1` | direction anchors — canvas, audience + `content_divergence` + `delivery_purpose` *(PPT only — omitted on non-PPT canvases, not written to the result)*, mode + visual_style | **Next** | writes `result.json` `{ stage: "tier1", status: "tier1-confirmed", <anchors> }`; the page does **not** close — it shows a "deriving…" state and polls `GET /api/recommendations` |-| `2` | design system — page count, color, icons, typography, formula policy | **Next** | writes `result.json` `{ stage: "tier2", status: "tier2-confirmed", <anchors + design system> }`; the page stays open and polls for Stage 3 |-| `3` | images and execution — image usage + generated-image style, generation mode, refine spec | **Confirm** | writes `result.json` `{ stage: "final", status: "confirmed", <all fields> }`, then shuts the page down |+| `"stage1"` | direction anchors — canvas, audience + `content_divergence` + `delivery_purpose` *(PPT only — omitted on non-PPT canvases, not written to the result)*, mode + visual_style | **Next** | writes `result.json` `{ stage: "stage1", status: "stage1-confirmed", <anchors> }`; the page does **not** close — it shows a "deriving…" state and polls `GET /api/recommendations` |+| `"stage2"` | design system — page count, color, icons, typography, formula policy | **Next** | writes `result.json` `{ stage: "stage2", status: "stage2-confirmed", <anchors + design system> }`; the page stays open and polls for Stage 3 |+| `"stage3"` | images and execution — image usage + generated-image style, generation mode, refine spec | **Confirm** | writes `result.json` `{ stage: "final", status: "confirmed", <all fields> }`, then shuts the page down || *(absent)* | legacy single-pass — every section on one page | **Confirm** | single final write (`status: "confirmed"`) — backward-compatible |-The AI launches Stage 1 (`--daemon --wait`), reads the stage-1 result, **re-derives** the design-system candidates from the user's actual anchors, overwrites `recommendations.json` with `"tier": 2`, and re-attaches with `--wait-only --wait-stage tier2`. After the Stage-2 result, it **re-derives** image and execution recommendations from the confirmed anchors + design system, overwrites `recommendations.json` with `"tier": 3`, and re-attaches with `--wait-only` for the final result. The page preserves earlier selections across transitions (single JS session). `GET /api/recommendations` is served `no-store` so polls see overwrites; on later stages the server folds already-confirmed choices from `result.json` back into the payload so a refresh / reopen re-initializes from the user's actual choices even though those sections are no longer rendered.+The AI launches Stage 1 (`--daemon --wait`), reads the stage-1 result, **re-derives** the design-system candidates from the user's actual anchors, overwrites `recommendations.json` with `"stage": "stage2"`, and re-attaches with `--wait-only --wait-stage stage2`. After the Stage-2 result, it **re-derives** image and execution recommendations from the confirmed anchors + design system, overwrites `recommendations.json` with `"stage": "stage3"`, and re-attaches with `--wait-only` for the final result. The page preserves earlier selections across transitions (single JS session). `GET /api/recommendations` is served `no-store` so polls see overwrites; on later stages the server folds already-confirmed choices from `result.json` back into the payload so a refresh / reopen re-initializes from the user's actual choices even though those sections are no longer rendered.### Input — `recommendations.json` (written by Strategist before launch)```json{+ "stage": "stage1","lang": "zh","recommend": {"canvas": "ppt169",@@ -191,11 +192,11 @@ The AI launches Stage 1 (`--daemon --wait`), reads the stage-1 result, **re-deri}```-The shape above is the **final** (Stage 3) result, carrying Stage 1 anchors, Stage 2 design-system fields, and Stage 3 image / execution fields. The intermediate **Stage 1** write carries only the anchor fields plus `"stage": "tier1"`, `"status": "tier1-confirmed"`; the AI reads it to re-derive Stage 2 and never treats it as the final confirmation. The intermediate **Stage 2** write carries anchors + design-system fields plus `"stage": "tier2"`, `"status": "tier2-confirmed"`; the AI reads it to re-derive Stage 3 and never treats it as final. A legacy single-pass write has no `stage` (or `stage: "final"`) and `status: "confirmed"`.+The shape above is the **final** (Stage 3) result, carrying Stage 1 anchors, Stage 2 design-system fields, and Stage 3 image / execution fields. The intermediate **Stage 1** write carries only the anchor fields plus `"stage": "stage1"`, `"status": "stage1-confirmed"`; the AI reads it to re-derive Stage 2 and never treats it as the final confirmation. The intermediate **Stage 2** write carries anchors + design-system fields plus `"stage": "stage2"`, `"status": "stage2-confirmed"`; the AI reads it to re-derive Stage 3 and never treats it as final. Legacy `tier1` / `tier2` results are accepted for old sessions but are no longer written. A legacy single-pass write has no `stage` (or `stage: "final"`) and `status: "confirmed"`.- Any option field may instead hold a **free-text custom string** (the user picked **Custom**); `color` / `typography` custom entries set `name: "custom"`. Image usage is not a custom string in new results: it is a source-id array, with free-text strategy captured in `image_notes`.- `image_ai_path` and `image_strategy` are omitted from `result.json` unless `image_usage` includes `ai`. Both are honored downstream as confirmed choices — and the page is only a convenience surface over the **canonical chat channel**: the same choices made in chat are honored identically when no `result.json` exists. `image_ai_path` drives the Step 5 generation path (`image-generator.md` §7 — `host-native` forces the host tool even when `IMAGE_BACKEND` is set); the chosen `image_strategy` candidate is locked verbatim by Strategist h.5 (no re-pick).-- After the user clicks the **final Confirm** (Stage 3, or single-pass), the page saves `result.json` and shuts the server down (auto-close). Stage-1 and Stage-2 **Next** keep the page open while it polls for the re-derived downstream stage. In the default flow, the first `--daemon --wait` returns on the stage-1 result, `--wait-only --wait-stage tier2` returns on the stage-2 result, and the final `--wait-only` returns on the final result; the AI reads each immediately — no extra chat confirmation is required. Chat confirmation remains the fallback when the page cannot be used. Either way, Step 4 ends with a `--shutdown` cleanup so a never-confirmed page cannot keep holding port 5050 ahead of the Step 6 live preview.+- After the user clicks the **final Confirm** (Stage 3, or single-pass), the page saves `result.json` and shuts the server down (auto-close). Stage-1 and Stage-2 **Next** keep the page open while it polls for the re-derived downstream stage. In the default flow, the first `--daemon --wait` returns on the stage-1 result, `--wait-only --wait-stage stage2` returns on the stage-2 result, and the final `--wait-only` returns on the final result; the AI reads each immediately — no extra chat confirmation is required. Chat confirmation remains the fallback when the page cannot be used. Either way, Step 4 ends with a `--shutdown` cleanup so a never-confirmed page cannot keep holding port 5050 ahead of the Step 6 live preview.## Scopediff --git a/skills/ppt-master/templates/decks/README.md b/skills/ppt-master/templates/decks/README.mdindex 1af14886..9900ce11 100644--- a/skills/ppt-master/templates/decks/README.md+++ b/skills/ppt-master/templates/decks/README.md@@ -48,7 +48,7 @@ Decks may include additional supporting sections (Voice & Tone, Icon Style, Layo## Fusion behavior at Step 3-When the user gives a deck path **alone**, Strategist locks all segments; Eight Confirmations narrows to deck-content fields (target audience / page count / outline / tone tweaks).+When the user gives a deck path **alone**, Strategist locks all segments; Strategist confirmation stage narrows to deck-content fields (target audience / page count / outline / tone tweaks).When the user gives a deck path **with** a brand path or layout path, identity / structure segments are overridden by the higher-priority source (brand wins on identity, layout wins on structure). See [`SKILL.md`](../../SKILL.md) Step 3 fusion table.diff --git a/skills/ppt-master/templates/layouts/README.md b/skills/ppt-master/templates/layouts/README.mdindex 008c972c..66e8d152 100644--- a/skills/ppt-master/templates/layouts/README.md+++ b/skills/ppt-master/templates/layouts/README.md@@ -1,6 +1,6 @@# Layout Templates-**Layout = structure-only template.** Captures canvas, page structure, page types, and SVG roster — but **no identity segment** (color / typography / logo / voice / icon style). Layered identity comes from `templates/brands/` or is decided per-deck in Strategist's Eight Confirmations. For full-identity replicas of specific PPTs, see [`templates/decks/`](../decks/) instead.+**Layout = structure-only template.** Captures canvas, page structure, page types, and SVG roster — but **no identity segment** (color / typography / logo / voice / icon style). Layered identity comes from `templates/brands/` or is decided per-deck in Strategist's confirmation stage. For full-identity replicas of specific PPTs, see [`templates/decks/`](../decks/) instead.Single source of truth for what layouts exist: [`layouts_index.json`](./layouts_index.json) (`layout_id → { summary, canvas_format, page_count, page_types }`). This README explains the kind; it does **not** enumerate layouts.diff --git a/skills/ppt-master/workflows/beautify-pptx.md b/skills/ppt-master/workflows/beautify-pptx.mdindex 08f192d1..9e8227fa 100644--- a/skills/ppt-master/workflows/beautify-pptx.md+++ b/skills/ppt-master/workflows/beautify-pptx.md@@ -136,7 +136,7 @@ If `images/image_manifest.json` does not exist because the source deck has no ex## 5. Beautify Plan — Recommend & Confirm-⛔ **BLOCKING**: the scope is not hard-coded — same spirit as the Eight Confirmations. Recommend each item below from what the deck actually contains (the Step 4 inventory), present the plan, and **wait for the user to confirm or adjust** before writing any spec. Chat is the canonical channel; the confirm UI below is the visual convenience surface over it for the palette + typography review (its result is honored identically to a chat reply).+⛔ **BLOCKING**: the scope is not hard-coded — same spirit as the Strategist confirmation stage. Recommend each item below from what the deck actually contains (the Step 4 inventory), present the plan, and **wait for the user to confirm or adjust** before writing any spec. Chat is the canonical channel; the confirm UI below is the visual convenience surface over it for the palette + typography review (its result is honored identically to a chat reply).This step has two halves:- **Visual re-confirm via the confirm UI** — the **full** Step 4 confirm page (below), seeded from the source so every targeted-confirmation field (canvas, mode, visual style, palette, icons, typography incl. body baseline, image strategy, generation mode) is **pre-filled with the inherited / source-derived default and left editable**. Beautify *recommends* keeping the source's identity, but never removes the user's place to override any field — you may choose not to change a value, but you must not deny the place to change it. This is also where the deck's text size is confirmed: `<stem>.identity.json` now carries size hints — `observed.sizes_pt` (the point sizes the deck actually renders at) and `theme.sizes` (the declared placeholder defaults) — so the `body_size` recommendation **follows the source's own font size** rather than a blind canvas default; the user still confirms or overrides it here.diff --git a/skills/ppt-master/workflows/failure-recovery.md b/skills/ppt-master/workflows/failure-recovery.mdindex 52e4f2cf..73132b5e 100644--- a/skills/ppt-master/workflows/failure-recovery.md+++ b/skills/ppt-master/workflows/failure-recovery.md@@ -15,8 +15,8 @@ Central recovery rules for common PPT Master failures. Route-specific workflow f| Failure point | Blocking | Automatic recovery | User intervention | Resume entry ||---|---:|---|---|---|| Confirm UI launch failure | No | Re-check `confirm_ui/result.json` once, then use chat fallback | No | `SKILL.md` Step 4 chat confirmation |-| Confirm UI wait timeout | No, if no final result yet | Re-check `result.json` once; keep server cleanup mandatory | Only if user still wants the page | Step 4 same tier or chat fallback |-| Confirm UI Tier 1 completed then interrupted | Yes until Tier 2 is written/confirmed | Read existing Tier 1 `result.json`, write Tier 2 recommendations, then `--wait-only` | Usually no | Step 4 Tier 2 write/wait |+| Confirm UI wait timeout | No, if no final result yet | Re-check `result.json` once; keep server cleanup mandatory | Only if user still wants the page | Step 4 same stage or chat fallback |+| Confirm UI Stage 1 completed then interrupted | Yes until Stage 2 is written/confirmed | Read existing Stage 1 `result.json`, write Stage 2 recommendations, then `--wait-only --wait-stage stage2` | Usually no | Step 4 Stage 2 write/wait || Missing final confirmation | Yes | None | User must confirm or change the values | Step 4 final confirmation || Formula rendering provider failure | No if fallback succeeds; yes if selected formulas remain missing | Provider fallback chain; otherwise mark affected rows manual only if acceptable | Only if rendered formula files are required and unavailable | Step 4 formula rendering / Step 7 image readiness gate || AI image generation failure | No | Retry once through the confirmed path, then mark row `Needs-Manual` | Only when missing files are required before export | Step 5 / Step 7 image readiness gate |@@ -57,7 +57,7 @@ Central recovery rules for common PPT Master failures. Route-specific workflow f| Last good state | Resume from ||---|---|-| Tier 1 confirmation exists, Tier 2 missing | Write Tier 2 recommendations, then `confirm_ui/server.py <project> --wait-only` |+| Stage 1 confirmation exists, Stage 2 missing | Write Stage 2 recommendations, then `confirm_ui/server.py <project> --wait-only --wait-stage stage2` || `design_spec.md` and `spec_lock.md` complete, split mode selected | [`resume-execute`](./resume-execute.md) || Images acquired but SVGs not started | `SKILL.md` Step 6 || SVGs complete and checker passed, notes missing | Step 6 Logic Construction |@@ -66,4 +66,4 @@ Central recovery rules for common PPT Master failures. Route-specific workflow f| Step 7.2 complete, PPTX not complete | Step 7.3 || Browser annotations saved after export | [`live-preview`](./live-preview.md) Step 2 |-**Default - resume at the owning failed step**: Do not restart Phase A or regenerate prior artifacts unless the owning source has changed.+**Default - resume at the owning failed step**: Do not restart the planning session or regenerate prior artifacts unless the owning source has changed.diff --git a/skills/ppt-master/workflows/index.md b/skills/ppt-master/workflows/index.mdindex 1bbf731f..140a846b 100644--- a/skills/ppt-master/workflows/index.md+++ b/skills/ppt-master/workflows/index.md@@ -19,8 +19,8 @@ Registry for standalone workflows referenced by `SKILL.md` and [`routing.md`](./| `beautify-pptx` | [`beautify-pptx.md`](./beautify-pptx.md) | Existing PPTX should be re-laid out while preserving slide count/order and wording 1:1 | Single source PPTX | Not for merge/split/drop/reorder or re-outline | Workflow Step 1 | Regenerated 1:1 deck through SVG pipeline | Blocks on source-slide mapping and final confirmation gates defined in the workflow || `create-template` | [`create-template.md`](./create-template.md) | Build a reusable layout/deck template from a PPTX or design reference | Design source or explicit template-creation request | Not a one-off generated deck | Workflow Step 1 | Template directory with `design_spec.md` and assets | Stops after template creation until user supplies the directory path to main Step 3 || `create-brand` | [`create-brand.md`](./create-brand.md) | Build/extract a reusable brand identity preset | Logo/site/branded deck/PDF or explicit brand setup request | Not a layout/deck template roster | Workflow Step 1 | Brand directory under `templates/brands/<id>/` | Blocks on brand evidence and validation defined in the workflow |-| `resume-execute` | [`resume-execute.md`](./resume-execute.md) | Continue Phase B for an existing project after split mode | Phase A artifacts exist | Do not re-run Phase A | Phase B entry | SVG generation, post-processing, export | Stops if `design_spec.md` or `spec_lock.md` is missing |-| `refine-spec` | [`refine-spec.md`](./refine-spec.md) | User explicitly opts into spec refinement before generation | Eight Confirmations completed | Do not enter by default | After Step 4, before Step 5/6 | Revised `design_spec.md` and `spec_lock.md` | Blocks for user review/approval before generation |+| `resume-execute` | [`resume-execute.md`](./resume-execute.md) | Continue an existing split-mode project | Planning-session artifacts exist | Do not re-run planning | Execution-session entry | SVG generation, post-processing, export | Stops if `design_spec.md` or `spec_lock.md` is missing |+| `refine-spec` | [`refine-spec.md`](./refine-spec.md) | User explicitly opts into spec refinement before generation | Strategist confirmation stage completed | Do not enter by default | After Step 4, before Step 5/6 | Revised `design_spec.md` and `spec_lock.md` | Blocks for user review/approval before generation || `verify-charts` | [`verify-charts.md`](./verify-charts.md) | Deck contains data charts that require coordinate calibration | SVG pages exist | Not needed for decks without data charts | Between Step 6 and Step 7 | Verified/fixed chart geometry | Blocks on chart coordinate errors || `customize-animations` | [`customize-animations.md`](./customize-animations.md) | User asks for object-level animation order/effect/timing | Target deck/SVG groups exist | Do not add per-element builds by default | Before or during export configuration | `animations.json` or equivalent validated config | Blocks if target objects cannot be identified || `native-enhance-pptx` | [`native-enhance-pptx.md`](./native-enhance-pptx.md) | Finished PPTX should keep content/layout stable while adding notes/audio/timing/transitions | Source PPTX exists | No SVG regeneration | Workflow Step 1 | Patched PPTX through OOXML | Blocks on source archive/validation failures |diff --git a/skills/ppt-master/workflows/refine-spec.md b/skills/ppt-master/workflows/refine-spec.mdindex c8fec3a8..715768a8 100644--- a/skills/ppt-master/workflows/refine-spec.md+++ b/skills/ppt-master/workflows/refine-spec.md@@ -1,12 +1,12 @@----description: Opt-in spec-refinement loop between the Eight Confirmations and Image/Executor. Triggered only when the user explicitly asks; the Strategist produces the full design spec, then HARD STOPS so the user can review and revise any part of it before the pipeline continues.+description: Opt-in spec-refinement loop between the Strategist confirmation stage and Image/Executor. Triggered only when the user explicitly asks; the Strategist produces the full design spec, then HARD STOPS so the user can review and revise any part of it before the pipeline continues.---# Refine Spec Workflow> Standalone, **opt-in** spec-review pass. The default pipeline writes `design_spec.md` + `spec_lock.md` and auto-proceeds. When the user explicitly asks to refine the spec, the Strategist produces the full spec first, then **stops** — the user reviews and revises any part of it (outline, color, typography, layout, image strategy, page rhythm, …) before any image generation or SVG work begins.-This workflow is **conditional**, same shape as the split-mode choice: it never fires on its own and the default path is unchanged. The Eight Confirmations settle design directions up front as abstract recommendations; this pass lets the user revise the **concrete spec** the Strategist produced from them. It is most valuable for a zero-background user, who can judge a finished spec far better than the up-front recommendations — and the spec's content outline (`§IX`) is usually what they most want to adjust.+This workflow is **conditional**, same shape as the split-mode choice: it never fires on its own and the default path is unchanged. The Strategist confirmation stage settles design directions up front as abstract recommendations; this pass lets the user revise the **concrete spec** the Strategist produced from them. It is most valuable for a zero-background user, who can judge a finished spec far better than the up-front recommendations — and the spec's content outline (`§IX`) is usually what they most want to adjust.## When to Run@@ -18,9 +18,9 @@ The user **explicitly asks** to refine / review / revise the spec before generat| "let me revise the spec, then continue" | "send me the spec to confirm, I'll edit it" || Any request to inspect/iterate the design spec before generation | "draft the full plan, I want to adjust it, then generate" |-**Default is OFF.** Strategist surfaces this option as one short opt-in line inside the Eight Confirmations bundle (see SKILL.md Step 4). No request → the spec is written in one go and the pipeline auto-proceeds as usual; this workflow never starts.+**Default is OFF.** Strategist surfaces this option as one short opt-in line inside the Strategist confirmation stage (see SKILL.md Step 4). No request → the spec is written in one go and the pipeline auto-proceeds as usual; this workflow never starts.-**Prerequisite**: the Eight Confirmations are settled (mode + visual style + the rest). This pass revises the spec the confirmations produced; it does not re-open the confirmation bundle itself.+**Prerequisite**: the Strategist confirmation stage is settled (mode + visual style + the rest). This pass revises the spec produced from that stage; it does not re-open the confirmation stage itself.---@@ -32,7 +32,7 @@ Run the default Strategist output exactly as SKILL.md Step 4 specifies: write `d## Step 2: ⛔ HARD STOP — present, discuss, and revise-Present the produced spec to the user and **wait for explicit revision or approval before doing anything else**. This is a conditional BLOCKING point that exists only on this opt-in path; the default pipeline keeps its "auto-proceed after the Eight Confirmations" discipline untouched.+Present the produced spec to the user and **wait for explicit revision or approval before doing anything else**. This is a conditional BLOCKING point that exists only on this opt-in path; the default pipeline keeps its "auto-proceed after the Strategist confirmation stage" discipline untouched.The user may revise **any part of the spec**, not just the outline — content outline, color, typography, layout, icon plan, image strategy, page rhythm. Discuss in **prose**; do not emit a scored rubric or per-axis grades (mechanical scorecards are against project convention). When useful, point out things worth a second look — but let the user drive.diff --git a/skills/ppt-master/workflows/resume-execute.md b/skills/ppt-master/workflows/resume-execute.mdindex 634b326b..39f3c8f5 100644--- a/skills/ppt-master/workflows/resume-execute.md+++ b/skills/ppt-master/workflows/resume-execute.md@@ -1,12 +1,12 @@----description: Phase B entry — resume PPT execution in a fresh chat after Phase A (SKILL.md Step 1-5) completed in a previous session. Reads project state from disk and runs Step 6 + Step 7 with no Phase-A context carry-over.+description: Execution-session entry — resume PPT execution in a fresh chat after the planning session (SKILL.md Step 1-5) completed in a previous chat. Reads project state from disk and runs Step 6 + Step 7 with no prior-chat context carry-over.---# Resume Execute Workflow-> Standalone Phase-B entry. Run when Phase A (SKILL.md Step 1–5) completed in a previous session and the user wants to continue with SVG generation + export. Loads project state from disk and runs Step 6 + Step 7 in a clean session.+> Standalone execution-session entry. Run when the planning session (SKILL.md Step 1–5) completed in a previous chat and the user wants to continue with SVG generation + export. Loads project state from disk and runs Step 6 + Step 7 in a clean session.-This workflow is **independent**: it owns Phase B starting from a fresh chat — no upstream conversation context required. By isolating SVG generation in its own session, the model gains 20–40K context headroom by not carrying Phase A's eight-confirmation dialogue, image search/fetch results, or Strategist references.+This workflow is **independent**: it owns the execution session starting from a fresh chat — no upstream conversation context required. By isolating SVG generation in its own session, the model gains 20–40K context headroom by not carrying the planning session's Strategist confirmation dialogue, image search/fetch results, or Strategist references.## When to Run@@ -18,13 +18,13 @@ The user opens a new chat and gives a phrase that names a project path and signa| "resume execution projects/<project_name>" | "resume execution projects/ppt169_joe_hisaishi" || Project path + any "继续 / 恢复 / 继续做 / 接着做" semantic | "把 projects/ppt169_joe_hisaishi 继续做完" |-**Prerequisite**: Phase A must have completed in the named project. Verified by file presence in Step 1; do NOT auto-trigger Phase A on missing state.+**Prerequisite**: the planning session must have completed in the named project. Verified by file presence in Step 1; do NOT auto-trigger planning on missing state.---## Step 1: Sanity check-Verify the project's Phase-A artifacts before doing anything else:+Verify the project's planning-session artifacts before doing anything else:| File / Directory | Required when | Reason ||---|---|---|@@ -33,7 +33,7 @@ Verify the project's Phase-A artifacts before doing anything else:| `<project_path>/images/` | `spec_lock images` references any image | Images must exist for embedding || `<project_path>/templates/` | `spec_lock page_layouts` / `page_charts` references any | Layout / chart SVGs needed for batch read |-If any required artifact is missing → report which one(s) and stop. Do NOT auto-fall-back into Phase A; the user must either complete Phase A in the original session or explicitly restart.+If any required artifact is missing → report which one(s) and stop. Do NOT auto-fall-back into planning; the user must either complete the planning session in the original chat or explicitly restart.---@@ -53,9 +53,9 @@ Then jump to `### Step 6: Executor Phase` and run the documented pipeline:- Speaker notes generation- Step 7: Post-processing & Export (`total_md_split` → `finalize_svg` → `svg_to_pptx`)-The fresh session pays the cost of re-reading references (~14K tokens) but earns back substantially more headroom by dropping Phase A's accumulated context. Net win in both window pressure and reasoning budget per page.+The fresh session pays the cost of re-reading references (~14K tokens) but earns back substantially more headroom by dropping the planning session's accumulated context. Net win in both window pressure and reasoning budget per page.-**Source materials**: Phase B is a fresh session; `<project_path>/sources/<file>.md` is NOT in context. The Executor SHOULD read the relevant `sources/` files when crafting per-page content — they hold the concrete facts, quotes, names, and details that turn skeleton outlines into substantive slides. `design_spec.md §IX` only carries the per-page intent; the source materials carry the texture. The Phase A → Phase B split is designed to free context budget precisely for this kind of high-quality enrichment.+**Source materials**: the execution session is fresh; `<project_path>/sources/<file>.md` is NOT in context. The Executor SHOULD read the relevant `sources/` files when crafting per-page content — they hold the concrete facts, quotes, names, and details that turn skeleton outlines into substantive slides. `design_spec.md §IX` only carries the per-page intent; the source materials carry the texture. The split-mode handoff is designed to free context budget precisely for this kind of high-quality enrichment.> Note: this workflow does NOT duplicate Step 6 / Step 7 content. SKILL.md is the authoritative procedure; resume-execute only adds the resumption entry (When to Run + Step 1 sanity check above) and the source-materials guidance above.diff --git a/skills/ppt-master/workflows/routing.md b/skills/ppt-master/workflows/routing.mdindex 6b396d99..ca5eab8f 100644--- a/skills/ppt-master/workflows/routing.md+++ b/skills/ppt-master/workflows/routing.md@@ -55,8 +55,8 @@ Route selection authority for PPT Master. Use this file before entering the main| Request shape | Trigger | Route | Preconditions | Output contract | Stop condition ||---|---|---|---|---|---|| Brand identity setup | Brand asset, brand site URL, branded PPTX/PDF, or explicit brand setup request | [`create-brand`](./create-brand.md) | Brand source exists or can be inspected | Brand preset under `templates/brands/<id>/` | Stop if no brand source or brand intent exists |-| Continue a split-mode project | "Continue generating `projects/<name>`" after Phase A | [`resume-execute`](./resume-execute.md) | Project has Phase A artifacts | Phase B SVG generation and export | Stop if required Phase A artifacts are missing |-| Refine spec before generation | User explicitly asks to refine/review/revise the spec before SVG work, or confirms `refine_spec: true` | [`refine-spec`](./refine-spec.md) | Eight Confirmations completed | Revised `design_spec.md` and `spec_lock.md` before Step 5/6 | Stop until user approves the refined spec |+| Continue a split-mode project | "Continue generating `projects/<name>`" after the planning session | [`resume-execute`](./resume-execute.md) | Project has planning-session artifacts | Execution-session SVG generation and export | Stop if required planning artifacts are missing |+| Refine spec before generation | User explicitly asks to refine/review/revise the spec before SVG work, or confirms `refine_spec: true` | [`refine-spec`](./refine-spec.md) | Strategist confirmation stage completed | Revised `design_spec.md` and `spec_lock.md` before Step 5/6 | Stop until user approves the refined spec || Data chart calibration | Generated deck contains data charts | [`verify-charts`](./verify-charts.md) between Step 6 and Step 7 | SVG pages exist | Calibrated chart coordinates before export | Stop on chart geometry errors until fixed || Object-level animation tuning | User asks for animation order, timing, effects, or object reveal behavior | [`customize-animations`](./customize-animations.md) | SVG groups / exported context exist | `animations.json` or validated animation config | Stop if requested target objects cannot be identified || Live preview / element selection / annotations | User mentions live preview, preview, visual check in browser, clicking/selecting an element, or applying browser annotations | [`live-preview`](./live-preview.md) | Project exists; for annotation apply, generated SVGs exist | Running preview service or applied annotations plus re-export | Stop only if project path or SVGs are missing |@@ -70,7 +70,7 @@ Route selection authority for PPT Master. Use this file before entering the main| User input | Route behavior ||---|---|| Explicit directory path containing a valid template `design_spec.md` | Enter main Step 3 template option |-| Bare template name, brand name, style label, or vague "use a template" | Do not trigger Step 3; treat as style input for Eight Confirmations |+| Bare template name, brand name, style label, or vague "use a template" | Do not trigger Step 3; treat as style input for Strategist confirmation stage || User asks "what templates exist?" | Answer as Q&A by listing indexed paths; do not advance the pipeline || Raw `.pptx` called a template | Route by §3, usually `template-fill-pptx`; never treat it as a Step 3 template path |