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
docs(roadmap): restructure around direction, compact shipped history
hugohe36f0e449ee027
diff --git a/docs/roadmap.md b/docs/roadmap.mdindex 37249c1a..c01d7ec2 100644--- a/docs/roadmap.md+++ b/docs/roadmap.md@@ -4,109 +4,49 @@----> PPT Master is a solo-maintained open source project, driven by **priority rather than fixed timelines**. This roadmap is here to align expectations: what's already shipped, what's under ongoing maintenance and evolution, and what's intentionally out of scope. Priorities shift with user feedback and real usage signals — no committed delivery windows.->-> **Where we are**: PPT Master has four explicit artifact routes. Generate PPTX uses constrained SVG → DrawingML to author newly designed slides; Create Template produces reusable Brand / Layout / Deck workspaces; Fill Native PPTX and Enhance Native PPTX preserve existing packages through scoped OOXML operations. Across all four, the core axis is **going ever deeper into PowerPoint's native object model** while keeping each route's editability, fidelity, and preservation trade-offs explicit.+> PPT Master is a solo-maintained open source project, driven by **priority rather than fixed timelines**. This roadmap aligns expectations: the direction of travel, what's in motion now, what may come when real demand shows, and what's intentionally out of scope. Priorities shift with user feedback and real usage signals — no committed delivery windows.----## Recent capability evolution+## Direction-Structural capability growth since March 2026. Single flags and incremental polish stay in the commit log.+The defining axis is **native depth**: author or preserve more of PowerPoint's own object model, behavior, and reusable structure, release after release — converging with PowerPoint itself. The [positioning charter](./project-positioning.md) states the full thesis; the [PowerPoint ↔ SVG Mapping Guide](./powerpoint-svg-mapping.md) records the current boundary honestly, feature by feature.-### 2026-03 — Native PPTX route takes shape+Today that axis is expressed through four explicit artifact routes: **Generate PPTX** authors newly designed slides through constrained SVG → DrawingML; **Create Template** produces reusable Brand / Layout / Deck workspaces; **Fill Native PPTX** and **Enhance Native PPTX** preserve existing packages through scoped OOXML operations.-- **Direct export to natively editable PPTX** — `svg_to_pptx` adds glow / rotate / text-decoration / stroke-linejoin; the full SVG → DrawingML chain becomes usable-- Chart / layout template JSON indexes ship, AI selection path connected--### 2026-04 — Pipeline at scale--- **Source-less generation**: the `topic-research` stage supports "topic only, no source files"-- **PPTX export step-change**: SVG clipPath → DrawingML picture geometry, marker → native arrows, output consolidated to `exports/`-- **Chart library expands to 70 templates + three icon libraries** (simple-icons / phosphor-duotone / brand-logo)-- **`spec_lock.md` machine-readable contract**: Strategist locks the spec, Executor rebuilds its read-only page-context projection before every page — cross-page consistency gets a real guarantee-- **Per-element animation support** + recorded narration / video export ([`workflows/stages/generate-audio.md`](../skills/ppt-master/workflows/stages/generate-audio.md)); the current behavior is opt-in, with element animation defaulting to `none`--### 2026-05 — Visual editing + AI image systematization--- **Live Preview enters the main pipeline** ([`workflows/stages/live-preview.md`](../skills/ppt-master/workflows/stages/live-preview.md)) — browser preview, click elements to write annotations, say "apply my annotations" and the AI rewrites that region (built on [@WodenJay](https://github.com/WodenJay)'s [PR #85](https://github.com/hugohe3/ppt-master/pull/85))-- **Create a reusable template workspace from a PPTX** ([`workflows/create-template.md`](../skills/ppt-master/workflows/create-template.md)) — read the supported theme / Master / Layout / asset facts that still exist in the package, then author or materialize a new validated Brand / Layout / Deck workspace without modifying the source PPTX-- **AI image three-dimension system** rendering × palette × type + [`strategist-image.md`](../skills/ppt-master/references/strategist-image.md) §2 lock, downstream consumes a fixed contract-- **AI image `hero_page` dual-track** — local insert + full-canvas hero image coexist-- **Brand identity preset subsystem** ([`workflows/create-template/create-brand.md`](../skills/ppt-master/workflows/create-template/create-brand.md)) — the Create Brand child workflow extracts and reuses brand palette / typography / logo / voice under the fixed Create Template entry-- **Visual self-review stage** ([`workflows/stages/visual-review.md`](../skills/ppt-master/workflows/stages/visual-review.md)) — rubric-based per-page check of AI-generated SVGs-- **AI image: Type concept boundary clarification** — Type is now narrowed to "the internal geometric skeleton of a local infographic block" (11 real skeletons); the four pseudo-types (hero / background / portrait / typography) fold back into `page_role: hero_page` plus four composition primitives (single-subject / portrait / typographic / atmospheric); hero_page text layering rule (visual keywords embedded, editable text via SVG overlay)-- **Brutalist AI newspaper example deck shipped** ([`examples/ppt169_brutalist_ai_newspaper_2026/`](../examples/ppt169_brutalist_ai_newspaper_2026/)) — first of the three P0 capability-backing demos: wall-to-wall small type + irregular columns + halftone monochrome + single-spot red + real native shapes; 10-page editorial annual report stressing text-position precision and cross-page consistency-- **Kubernetes Blueprint example deck shipped** ([`examples/ppt169_kubernetes_blueprint_2026/`](../examples/ppt169_kubernetes_blueprint_2026/)) — second of the three P0 capability-backing demos: isometric technical-drawing aesthetic + blueprint cyan/amber palette + hand-authored SVG geometry (no raster images) + custom drawing-in animation; 10-page Kubernetes architecture walkthrough stressing geometric shape generalization and chart-structure extensibility-- **AI image `custom` escape hatch** — `rendering` / `palette` / hero composition each accept `custom` + a one-paragraph `*_behavior` prose, replacing the false "default to vector-illustration / cool-corporate" fallback; end-to-end contract spans [`image-renderings/_index.md`](../skills/ppt-master/references/image-renderings/_index.md) §1.5, [`image-palettes/_index.md`](../skills/ppt-master/references/image-palettes/_index.md) §2, [`strategist-image.md`](../skills/ppt-master/references/strategist-image.md) §2 hard-rule (≤1 custom per dimension; one candidate may carry both), spec_lock fields, and Image_Generator Step 2 consumption branch-- **Template architecture: three-kind consolidation** ([`docs/templates-architecture.md`](./templates-architecture.md)) — brand / layout / deck split into three independent dirs with per-kind schemas + segment-level fusion + git-style conflict resolution; [Generate PPTX Step 3](../skills/ppt-master/workflows/generate-pptx.md#step-3-template-option) dispatches per `kind`, trigger rule remains "explicit path only"-- **Pattern fill PPTX safety net** — `svg_quality_checker.py` warns when a referenced `<pattern>` lacks `data-pptx-pattern` and the converter will use its compatible `ltUpDiag` fallback; it still rejects `patternTransform` and values outside OOXML `ST_PresetPatternVal`. [`native-data-interface.md` §1](../skills/ppt-master/references/native-data-interface.md#1-pattern-fill--pattern-with-pptx-preset-annotation) documents the closed preset enum and color resolution from importer metadata or child paint-- **LaTeX math formula rendering shipped** ([`scripts/latex_render.py`](../skills/ppt-master/scripts/latex_render.py)) — Strategist locks one of three policies (`mixed` / `render-all` / `text-only`) inside the Typography confirmation and writes an explicit `images/formula_manifest.json`; the renderer walks a codecogs → quicklatex → mathpad → wikimedia fallback chain and emits transparent PNGs that land in §VIII as `Acquire Via: formula` / `Status: Rendered` rows; formula-heavy decks (academic / engineering / educational) finally have a native rendering route. Formula selection is a Strategist decision — the renderer never scans source files for `$...$` markers-- **Live preview direct editing — L1 / L2 / L3** ([`workflows/stages/live-preview.md`](../skills/ppt-master/workflows/stages/live-preview.md)) — the browser editor gains deterministic in-place edits with no AI round-trip: text content (L1), presentation attributes like fill / stroke / font-size (L2), and on-canvas geometry (L3) — drag a selected element to move it, arrow-key nudge (`Shift` = 10px), multi-select, plus a right-click overlap picker for stacked shapes. Edits stage with `Ctrl+Z` undo + coalescing and write to `svg_output/` on **Apply changes**; moves persist through finalize / export (moved text frames, promoted multi-line tspans, repositioned icons all reproduce in the PPTX). Re-export stays chat-driven; on-canvas resize handles are not yet implemented (resize via the geometry inputs)--### 2026-06 — Mode & visual-style dual catalogs + PPTX intake & content-strategy expansion--- **Reuse a native PPTX's design → refill content route** ([`workflows/template-fill-pptx.md`](../skills/ppt-master/workflows/template-fill-pptx.md)) — when a user supplies an existing `.pptx` plus new material / a topic and asks to "reuse this deck's design / fill the content back in", the Fill Native PPTX route edits the PPTX directly and never enters the SVG generation pipeline. Output stays natively editable (it reuses the original slides' shapes / layouts, not a screenshot refill); it isolates private parts on reuse, exposes chart data, and runs capacity checks. Trigger follows the template rule — only on an explicit ask to reuse an existing deck — and it deliberately does not reflow / add pages / swap images (that's the Generate PPTX route). Distinguished from Non-goals #53 below-- **Three executors retired → mode + visual-style dual catalogs** ([`references/modes/`](../skills/ppt-master/references/modes/) + [`references/visual-styles/`](../skills/ppt-master/references/visual-styles/)) — the old three `executor-*.md` (general / consultant / consultant-top) entangled domain · audience · persuasion · narrative on one axis; split into two orthogonal catalogs (following the `image-renderings` pattern: flat dir + `_index` + on-demand read + Strategist locks one). **mode** = narrative skeleton (`pyramid` / `narrative` / `instructional` / `showcase`; consultant + top merge into pyramid since their narrative core is identical); **visual-style** = SVG layout aesthetic (`swiss-minimal` / `editorial` / `soft-rounded` / `dark-tech`, each paired with an image-rendering, **zero HEX** — color truth stays in confirmation e + image-palettes). Strategist `§d` locks `mode` + `visual_style` independently into `spec_lock`; Executor loads the two locked files; any mode × any style. Render coordinates stay in `templates/charts/`-- **Prompt constraint-strength decoupling** ([`docs/rules/prompt-style.md`](rules/prompt-style.md) §4) — three explicit strength tiers — rule (`Hard rule` / `Forbidden`) / default (`Default — … may override`) / reference (`Reference — not a constraint`) — plus an "objective failure vs taste" test and a checker boundary, so the model can tell "must keep vs may deviate" at a glance; the visual-style catalog is Reference-strength throughout-- **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 Cangzhuo / Li Ziqi) · `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-structured.md`](../skills/ppt-master/references/executor-structured.md) §1 re-skins chart/layout template gradients-shadows-fills to the deck's visual_style (templates supply structure, not skin; mirror templates retain their declared prototype paint 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/stages/refine-spec.md`](../skills/ppt-master/workflows/stages/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. Template application is now AI-authored from the actual workspace and current content; the page does not expose reuse/adherence controls. Internal strict/adaptive values remain available to the structured exporter. The page shares port `5050` with Live Preview and writes authoritative user-facing choices to `result.json`.-- **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 profile shipped** ([`workflows/profiles/beautify-pptx.md`](../skills/ppt-master/workflows/profiles/beautify-pptx.md)) — a main-generation profile that mirrors `template-fill`: template-fill reuses a deck's design with new content, while 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 it remains the Generate PPTX route (`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 normal profile), does not guarantee coordinate-level paste-back, and combo / dual-axis / waterfall charts lose the un-captured plots--- **Multi-deck PPTX intake + `analysis/` source-name prefixing** — a main-pipeline project can now combine several source decks: each writes its own `<stem>.identity.json` / `<stem>.slide_library.json`, with every deck's digest inlined under `decks[]` in the single index `source_profile.json` (preserving the "Strategist must-read `source_profile.json`" single-entry contract — one entry for a one-deck project, several for a combined one; re-importing the same stem replaces its entry). `beautify` / `template-fill` stay single-deck (1:1) and read their own `<stem>.*` artifacts--- **Material divergence (a Stage-1 free-text source-treatment field)** — a main-pipeline content-strategy question added to the delivery / source-treatment section: the user states in their own words how closely to follow the source vs how freely to reshape it (blank = balanced default). Deliberately not a fixed set of options, not recommended from analyzing the source, and not coupled to page count — it is purely the user's stated intent. However freely they ask, facts **stay sourced** — reorganize / reframe / expand / connect what is in the source, never import facts from outside it (that is the `topic-research` stage's job). The Strategist reads the prose when authoring the §IX outline and records it in `design_spec §I`; it is **not** written to `spec_lock` (the Executor never reads it). `mode` and divergence are orthogonal. Beautify seeds verbatim preservation and surfaces the field as locked/read-only; template-fill does not surface it because that route bypasses this confirmation flow--- **A batch of default-behavior and intake standardizations** — per-element entrance animation now defaults off (page transition `fade` only; element builds are opt-in via `-a auto` / `animations.json`), removing the auto-cascade "AI deck" tell; per-project `icons/` copies chosen icons into the project at selection time and embeds project-first; `analysis/` is established as the canonical must-read layer for machine-extracted facts (PPTX intake bundle + `image_analysis.csv`); the main pipeline treats a source deck's identity (palette / fonts / layout) as **reference, not constraint** (inherit or redesign by Strategist judgment, defaulting to fresh design); the confirm page gains a custom color input--- **AI illustration sheet → sliced spot-illustration pipeline** ([`scripts/slice_images.py`](../skills/ppt-master/scripts/slice_images.py)) — when a deck wants several spot illustrations of one family, instead of one AI call per cell it **generates a single multi-cell illustration sheet** (one call locks the whole set's style / palette and costs far less than per-cell generation), then `slice_images.py` deterministically cuts it on an `RxC` grid into individual element files; `--trim` tight-crops to each cell's content bounding box and `--alpha` knocks out the flat background, so each element lands as a **transparent cutout** on a differently-colored page rather than a visible box. Two §VIII resource rows make this explicit: one `ai` Illustration Sheet (generated but never placed directly, kept out of `spec_lock.md` images) plus one `slice` element row per cell (actually placed, listed in spec_lock); Step 5 generates the sheet, slices it, and re-runs `analyze_images`, and the Step 7 readiness gate handles the offline case by listing the sheet + element targets for the user to place and re-slice. Whether the deck uses spot illustrations is a Strategist judgment within the `image_usage` source boundary (not a separate confirm field; `image_usage: none` always wins over illustration intent); users never see the internal sheet/slice implementation. `svg_quality_checker` gains matching validation, and [`image-layout-patterns.md`](../skills/ppt-master/references/image-layout-patterns.md) adds image-led promotional layout patterns--- **Spot illustrations: from "can be sliced" to "used well, as a system"** — a decision layer on top of the slice pipeline plus edge-quality hardening, so a deck actually uses illustration well rather than merely being able to generate it. ① **Slice quality**: `slice_images.py` `--alpha` now uses a feathered soft mask (anti-aliased + 1px defringe), samples the background from a 2px border-ring median, and measures color distance as the max channel difference; `svg_quality_checker` verifies generated `slice` rows have their files. ② **Trigger lean tied to `visual_style`**: each style carries a `core` / `supportive` / `sparse` illustration propensity ([`visual-styles/_index.md`](../skills/ppt-master/references/visual-styles/_index.md) `Illus.` column + each file's §6) — `core` styles recommend use by default, `sparse` default to none; precedence is `image_usage: none` → explicit user intent (overrides either way) → style propensity → none. ③ **Motif through-line**: when a deck leans into illustration, the cover anchor / section dividers / body spots are drawn from one motif family (shared h.5 rendering + palette + subject world) so the deck reads as one designed system rather than scattered spots; the AI motif generates only when `image_usage` includes `ai`, while `provided` / `web` reuse only assets that already form one family. ④ **Illustration-roles decision map**: a Strategist §h navigation table (role × when × mechanism × source — spot / hero anchor / section divider / atmospheric background / motif). Throughout: no quota, and no turning taste into a checker rule (a same-tiles placement detector was evaluated and declined, left to §4.3 placement prose + execution judgment)--- **Image transform-matrix fidelity end-to-end + host-native generation path** — `svg_to_pptx`'s DrawingML picture export now honors the SVG transform matrix (rotation / skew / compound transforms no longer shift or collapse to the origin under nested `<g transform>`), extending the cross-four-renderer position-fidelity axis to the raster layer; `image_gen.py` adds a host-native generation path that routes through the host's own image-generation capability when present. Both are fixes / incremental hardening — details in the commit log--- **Web image sourcing → best-match + reviewable + manual replace** — web rows no longer silently download a candidate pool; the default downloads **only the best match**, with the pool demoted to an opt-in `--save-candidates` escalation (default 4). Every download writes a ≤1024px review copy under `images/.review/` (placed / promoted asset stays full-resolution). Suitability review is **model-agnostic**: a multimodal model reads the review copy, a non-multimodal one hands the `source_page_url` to a human — no assumption of vision. New `image_search.py --from-url <url>` downloads any human-chosen image URL into the target (recorded `license_tier: manual`, inheriting page context) as the universal manual-replacement path; `--promote` now recomputes the credit from the chosen candidate (never the replaced image's). It all runs inside Step 5 and never halts the pipeline (a poor fit becomes `Needs-Manual` + placeholder). Web search is positioned as a placeholder-grade fallback that does **not** guarantee quality — high quality comes from AI generation or picking an image by hand and swapping it in--- **Web image entity-safety gate (exact subjects no longer lose to a "hi-res wrong image")** — building on the row above: web candidates now pass a `required_terms` entity gate that blocks metadata-relevant-but-wrong-subject images (a polished Rome monument winning a "Chongqing landmark" row). Groups are ANDed, `A|B` gives intra-group aliases (for example `Chongqing|Chungking`), and matching is lowercased / separator-normalized / space-compacted so multi-word and CJK names match; an entity hit counts as the relevance signal (a zero query-token overlap no longer rejects, so a CJK-titled landmark passes under an English query — the old reject still applies when no `required_terms` are set). Pixel area drops from a dominant score (cap 5000) to a tie-breaker (cap 1500) plus a title-hit boost, so entity accuracy and relevance outrank raw resolution. CLI `--require-terms` (repeatable, comma / `|`), batch `required_terms`, and `--from-url` all inherit it, recorded in `image_sources.json` for audit; a toothless gate (weak `required_terms`) emits a warning. It is a **metadata gate** paired with the `.review` visual check, not a visual classifier--- **Native PPTX export caps image media size** — keep generated decks editable without embedding huge source images: new native image sizing modes — `cap` (default) limits only oversized source images to a max dimension, while `display` sizes from the rendered SVG box for more aggressive compression; native export preserves full embedded pixels while SVG/PPT display cropping stays as editable picture-crop metadata; `finalize_svg` keeps its existing slice/meet behavior and adds default rendered-size downsampling for compact SVG snapshots. Docs land the `cap` / `display` modes and the `--no-image-optimize` escape hatch--### 2026-07 — Staged confirmation UI + native objects + motion hardening--- **Step 4 confirmation gate restaged into a three-pass wizard + visual previews** — the single Eight-Confirmations gate became a dependency-ordered flow in one browser session: open communication contract → complete deck solution → resources / production. Communication purpose is free prose that can preserve several goals and their relationship; common purpose names are hints, never a `primary_job` selector. Every editable Stage-1 prose box contains a recommendation rather than a required answer: confirmation persists its current value exactly, so a user-cleared field stays empty through the final result. Stage 2 keeps mutually dependent visual choices together as coordinated safe / shifted / bold directions, while Stage 3 contains no aesthetic choices. Generated images confirm rendering only and inherit the selected deck colors directly. Each downstream stage is derived from the user's **actual** confirmed upstream answers. `recommendations.json` uses the canonical `stage` selector (`tier` remains read-only compatibility input); the chat fallback mirrors the same staged order-- **Independent AI-image palette confirmation retired** — the May rendering × palette × type model remains readable for legacy projects and comparison assets, but new recommendations / locks no longer author `image_palette`. Stage 2 confirms generated-image rendering inside each coordinated deck direction; Image_Generator consumes the selected deck color roles directly. This removes a duplicate color decision and prevents same-stage palette / HEX mismatch--- **Transitions and entrance-animation export hardened against silent downgrade** — the current defaults stay `fade` / 0.4 s for page transitions and `none` for element entrances; object motion remains opt-in through `-a` or `animations.json`. Unknown effects / Start modes, non-finite or out-of-range timing, invalid order values, and missing slide/group references now fail instead of falling back to `fade`, another Start mode, or inherited timing. Export reads the candidate PPTX back before replacement and validates root timing placement, unique time-node IDs, live shape targets, effect/duration/Start semantics, and narration timing merges. Direct-PPTX routes preserve existing object animation rather than translating it into the generated animation model. Microsoft PowerPoint is the primary behavior-validation target; other presentation apps remain compatibility targets without an identical-motion guarantee+----- **`--native-charts-and-tables` hardened from dormant marker to usable opt-in** — the narrow native Chart/Table exception (see Non-goals below) now emits charts and pure-text tables that keep the deck's own design instead of snapping to PowerPoint's white default theme. Classic native charts write explicit chart-area / plot-area / axis-line / gridline / label-text colors — inferred from the visible SVG fallback (largest panel `<rect>` → background, fallback text → labels, fallback strokes → axis/grid) or overridden per `style` (`chart_area_fill` / `plot_area_fill` / `text_color` / `axis_color` / `grid_color`, `"none"` for transparent); the color parser normalizes named CSS colors, `#RGB` shorthand, and `rgb()` / `rgba()` to OOXML hex; bar/column series disable negative-value fill inversion so negative bars keep their series color. The activated export is named `<name>_<ts>_native_charts_tables.pptx` to tell it apart from the default shape-based export. **The default route is unchanged** — charts/tables still export as editable SVG-derived DrawingML shapes for cross-renderer fidelity; PowerPoint-native Chart/Table objects add a data source and object-specific editing model and stay the deliberate opt-in trade documented under Non-goals below+## In progress / Next-- **Native package structure + mode-specific template creation** — deck/layout-template SVG pages declare their final Master/Layout identity while they are authored. Fixed Master/Layout visuals are root-level atoms, reusable slots are bounded top-level groups with real carriers or an explicit composite proxy, and zero-slot Layouts remain valid. Structured export deterministically compiles only that contract and performs final package read-back; it never promotes repeated chrome or infers placeholders. Flat free-design/brand-only export keeps all content Slide-local while replacing the stock Office scaffold with one clean project-owned Master and Blank Layout plus a deck-named current-lock theme. It removes title/body and other stock content, prunes unused Layouts, and retains only the standard date/footer/slide-number capability hooks. Both generation modes write the locked title plus a deterministic nine-level body hierarchy into Master `p:txStyles`, preserving paragraph and bullet settings. `standard` / `fidelity` author new SVG rosters and new Master/Layout systems without preserving or distilling source topology. `mirror` materializes a new workspace from the complete set of supported facts actually present in a native source package—including definition-only prototypes for unused Layouts—without semantic synthesis; fixed-layer source groups are mechanically expanded into direct atoms. Lossless imports stay immutable in analysis as payload backing, while `authoring-svg/` plus its manifest form the editable template-creation IR. Library and project outputs share required `templates/`, optional `images/` / `icons/`, and optional on-demand review output in `exports/`; only library registration differs. Legacy baseline/template/preserve SVG packages are not migrated: use them only as visual reference while `create-template` authors a new workspace, then generate a new PPTX from that workspace. Raw one-off PPTX filling remains `template-fill-pptx`.+Actively underway or up next — no committed timeline.-- **Native-preset authoring guidance strengthened across roles** — using a real PowerPoint preset (`prstGeom`, with adjustment handles and the non-flat-card look) is reaffirmed as the default for stock geometry (block arrows, chevrons, banners, callouts, flowchart nodes, stars), not an exception. New pages and project-owned templates use the compact atomic `<g>` emitted by `preset_shape_svg.py`: the group owns the preset metadata and base paint once, while its direct visible paths are registry-derived layers with only necessary per-layer paint overrides. The authoring form has no hidden carrier, preview wrapper, or stored fingerprint; it is helper-only, and any change to preset, frame, adjustments, or paint regenerates the whole group instead of editing paths or metadata. A matching compact fragment may therefore live in `templates/charts/` when it expresses an intentional stock shape. The Executor still decides by object intent at draw time rather than scanning finished paths, and the paint boundary remains narrow: gradient fill/stroke or pattern fill stays ordinary SVG. PPTX import and `mirror` keep the separate expanded lossless representation. The Strategist continues to prime likely pages with a non-destructive `native-preset candidate` note; the Executor selects the exact preset, frame, and paint+- **Calibrate the recently landed systems on real decks** — multi-deck intake, the material-divergence field, the spot-illustration system, and structured template creation have all shipped; what they need now is real-usage signal, not more mechanism. No pre-emptive thresholds or quotas.+- **Prompt slimming** — compress per-role prompt token footprint and improve cache hit rate without sacrificing quality. This is the indirect cost/speed lever; the boundary with quality-sacrificing speedups is drawn under Non-goals.----## In progress / Next+## Future directions — signal-driven-Directions actively underway or up next, with no committed timeline.+Candidates already evaluated as "worth doing when real demand shows", listed so the intent is public. None is a commitment.-- **Real-usage calibration of multi-deck intake and material divergence (just landed)** — multi-deck combined intake (`<stem>` prefixing + a `decks[]` merge index) and the Stage-1 material-divergence free-text field both shipped (see "2026-06" above); next is calibration from real usage: same-stem collisions across decks currently let the later one overwrite the former — whether dedupe / numbering is needed awaits signal; and whether the Strategist reads the free-text intent accurately, and whether the "facts stay sourced" line holds when the user asks for free reshaping, awaits real generation. Neither gets a mechanical threshold added pre-emptively-- **Real-deck calibration of the illustration capability (mechanism + deployment layer, just landed)** — the slice pipeline, the edge-quality hardening, and the decision layer (style propensity / motif through-line / roles map — see "2026-06" above) have all shipped; next is calibration from real usage: whether one sheet sliced into many cells holds style / palette consistency and how robust the `--alpha` soft mask is against irregular composition inside a cell, the offline readiness-gate manual-place + re-slice experience, whether propensity flips the styles it should, whether the motif reads as a designed system rather than over-decoration on real decks, and whether the source boundary holds (no silent AI on `provided` / `web`). No mechanical threshold / quota added pre-emptively; if same-size tiles really recur, consider a narrower lint then-- **Native drop-shadow on authored presets — deferred as a separate enhancement** — the compact authored-preset `<g>` is a closed semantic atom and currently admits only its registered preset metadata, base paint, and direct registry paths; `filter` / effect metadata is not part of that authoring contract. The converter may expand the atom into its established carrier/preview transport internally, but that implementation detail is not a source-SVG API. Imported PPTX and `mirror` may retain supported source effects in the separate expanded lossless representation; that does not widen new authoring. Until a precise preset-effect contract and matching checker coverage exist, a stock shape that needs a shadow conservatively stays ordinary SVG. This remains signal-driven rather than an implied capability-- Otherwise: the mode / visual-style validation and calibration is settled (see "2026-06" above) — the structure (5 modes + 18 visual-styles + custom) is locked, the four adjacent pairs are consolidated into one Close-calls table, and the four calibration tightenings have landed. Future direction is driven by real usage signal and feedback; long-running improvements are under "Ongoing maintenance" below, and evaluated-out directions under "Non-goals"+- **Keep closing the native-coverage gaps** recorded in the [mapping guide](./powerpoint-svg-mapping.md) — release after release, move more "SVG-only" cells toward native PowerPoint structure and behavior.+- **Effects on authored preset shapes** (e.g. a native drop-shadow) — waits for a precise preset-effect contract plus checker coverage; until then, a stock shape that needs a shadow conservatively stays ordinary SVG.+- **Hyperlink authoring in generated decks** — hyperlinks already present in source decks survive conversion today; letting the Strategist author new links waits for demand.+- **Picture slide backgrounds as native background fill** — solid/gradient page backgrounds already export as PowerPoint-native slide backgrounds; the picture case is demand-driven.----## Ongoing maintenance directions+## Shipped milestones-Long-running improvements with no committed timeline. Only real directions are listed; specific fixes / single flags go to the commit log.+One line per month. Full detail lives in the [release notes](https://github.com/hugohe3/ppt-master/releases) and the commit log.-- **Prompt slimming** — compress per-role prompt token footprint and improve cache hit rate without sacrificing quality, for indirect cost / speed gains. Complements "Pure speed optimization" below: indirect optimization yes, quality-sacrificing speedups no.+| When | Theme |+|---|---|+| 2026-03 | **Native PPTX route takes shape** — the SVG → DrawingML chain becomes usable; chart/layout template indexes ship |+| 2026-04 | **Pipeline at scale** — topic-only generation, 70 chart templates + three icon libraries, the `spec_lock` cross-page consistency contract, per-element animation and narration/video export |+| 2026-05 | **Visual editing + AI-image systematization** — Live Preview with deterministic in-place editing (built on [@WodenJay](https://github.com/WodenJay)'s [PR #85](https://github.com/hugohe3/ppt-master/pull/85)), template workspaces from PPTX, the rendering × palette × type image system, LaTeX formula rendering |+| 2026-06 | **Mode & visual-style dual catalogs + intake expansion** — 5 narrative modes × 18 visual styles (+ `custom`), content-faithful beautify profile, multi-deck intake, spot-illustration pipeline, web-image quality gates, source-conversion fidelity gains (caption recognition from [@suay1113](https://github.com/suay1113)'s [PR #191](https://github.com/hugohe3/ppt-master/pull/191), hyperlink preservation distilled from [@ZhaoZuohong](https://github.com/ZhaoZuohong)'s [PR #155](https://github.com/hugohe3/ppt-master/pull/155)) |+| 2026-07 | **Positioning charter + native masters & layouts + token efficiency** ([v4.0.0](https://github.com/hugohe3/ppt-master/releases/tag/v4.0.0)) — three-pass staged confirmation UI, real `p:sldMaster` / `p:sldLayout` export, `--native-charts-and-tables` opt-in, motion-export hardening, chart template library compacted |---@@ -130,7 +70,7 @@ The Generate PPTX route is built around full control of newly authored shapes, tPixel-fidelity across the four renderers (PowerPoint / Keynote / LibreOffice / WPS) is the project's spine. Switching the default route to native PowerPoint charts breaks that — the same PPTX renders different chart layouts across renderers. Charts as SVG is **by design**, not a capability gap.-The narrow exception is the `data-pptx-replace-with` marker: supported data charts and pure text-grid tables carry a PowerPoint-native Chart/Table replacement payload at generation time, and exporting with `--native-charts-and-tables` activates it for users who deliberately trade cross-renderer fidelity for a data-backed object and its chart/table-specific editing model — the activated objects preserve the deck's chart-area / plot / axis / gridline / label colors and native table formatting rather than snapping to PowerPoint's default theme (see 2026-07 above). The default export path and editable SVG-derived shape system are unchanged.+The narrow exception is the `data-pptx-replace-with` marker: supported data charts and pure text-grid tables carry a PowerPoint-native Chart/Table replacement payload at generation time, and exporting with `--native-charts-and-tables` activates it for users who deliberately trade cross-renderer fidelity for a data-backed object and its chart/table-specific editing model — the activated objects preserve the deck's chart-area / plot / axis / gridline / label colors and native table formatting rather than snapping to PowerPoint's default theme (see the [v4.0.0 release notes](https://github.com/hugohe3/ppt-master/releases/tag/v4.0.0)). The default export path and editable SVG-derived shape system are unchanged.### uv as default / required dependencydiff --git a/docs/zh/roadmap.md b/docs/zh/roadmap.mdindex 94a6b466..2eb27828 100644--- a/docs/zh/roadmap.md+++ b/docs/zh/roadmap.md@@ -1,112 +1,52 @@-# Roadmap+# 路线图-[English](../roadmap.md) | [中文](./roadmap.md)+[English](../roadmap.md) | [Chinese](./roadmap.md)----> PPT Master 是单人维护的开源项目,按**优先级**而非时间表推进。这份 roadmap 用来统一对外预期:已经做了什么、在持续维护演进什么、暂时不打算做什么。优先级会随用户反馈和实际使用信号调整,不承诺时间窗口。->-> 项目当前有四条显式产物路线:Generate PPTX 通过受约束的 SVG → DrawingML 创作新页面;Create Template 产出可复用 Brand / Layout / Deck 工作区;Fill Native PPTX 与 Enhance Native PPTX 则通过受控 OOXML 操作保留既有包结构。四条路线共同围绕一个主轴:**不断深化对 PowerPoint 原生对象模型的支持**,同时把每条路线的可编辑性、保真度与保留边界说清楚。+> PPT Master 是一个由个人维护的开源项目,按**优先级而非固定时间表**推进。这份路线图用来对齐预期:项目往哪个方向走、当下在做什么、哪些事等真实需求出现再做、哪些明确不做。优先级会随用户反馈和真实使用信号调整——不承诺交付时间窗口。----## 近期能力演进+## 方向-这里记录自 2026 年 3 月以来的结构性能力演进;单 flag 与增量优化看 commit log。+项目的主轴是**原生深度**:逐版本创作或保留更多 PowerPoint 自身的对象模型、行为与可复用结构——持续向 PowerPoint 本身靠拢。完整论述见[项目定位章程](./project-positioning.md);[PowerPoint ↔ SVG 映射指南](./powerpoint-svg-mapping.md)逐特性诚实记录当前边界。-### 2026-03(真原生 PPTX 路线成型)+这条主轴今天体现为四条显式产物路线:**Generate PPTX** 通过受约束的 SVG → DrawingML 创作全新设计的页面;**Create Template** 产出可复用的 Brand / Layout / Deck 模板工作区;**Fill Native PPTX** 与 **Enhance Native PPTX** 通过限定范围的 OOXML 操作保留既有文件包。-- **直接导出原生可编辑 PPTX** — `svg_to_pptx` 补齐 glow / rotate / text-decoration / stroke-linejoin,整条 SVG → DrawingML 链路开始可用-- 图表 / 布局模板 JSON 索引上线,AI 选型路径打通--### 2026-04(管线规模化)--- **无源生成**:`topic-research` 阶段支持「只给主题、不给源文件」-- **PPTX 导出质变**:SVG clipPath → DrawingML picture geometry、marker → 原生箭头、输出归集到 `exports/`-- **图表库 70 个 + 图标三库**(simple-icons / phosphor-duotone / brand-logo)-- **`spec_lock.md` 机器可读契约**:Strategist 锁定后 Executor 每页强制重读,跨页一致性有了保证-- **元素级动画能力** + 旁白音频 / 视频导出([`workflows/stages/generate-audio.md`](../../skills/ppt-master/workflows/stages/generate-audio.md));当前行为为按需开启,元素动画默认 `none`--### 2026-05(视觉编辑 + AI 图系统化)--- **Live Preview 进入主流程**([`workflows/stages/live-preview.md`](../../skills/ppt-master/workflows/stages/live-preview.md)) — 浏览器实时预览 + 点选元素写要求 + 「apply my annotations」让 AI 重做该区域(基于 [@WodenJay](https://github.com/WodenJay) [PR #85](https://github.com/hugohe3/ppt-master/pull/85))-- **从原生 PPTX 创建新模板工作区**([`workflows/create-template.md`](../../skills/ppt-master/workflows/create-template.md)) — 读取 OOXML 中实际存在且受支持的主题 / 母版 / 版式 / 资源事实,物化为新的模板合同;不修改来源 PPTX-- **AI 图三维系统** rendering × palette × type + [`strategist-image.md`](../../skills/ppt-master/references/strategist-image.md) §2 锁定,下游消费固定契约-- **AI 图 `hero_page` 双档** — 局部插图 + 整页主角图共存-- **品牌身份预设子系统**([`workflows/create-template/create-brand.md`](../../skills/ppt-master/workflows/create-template/create-brand.md)) — 固定 Create Template 入口下的 Create Brand 子工作流,提取并复用品牌色板 / 字体 / Logo / 语调-- **视觉自检阶段**([`workflows/stages/visual-review.md`](../../skills/ppt-master/workflows/stages/visual-review.md)) — 按 rubric 逐页自查 AI 生成的 SVG-- **AI 图 Type 概念边界澄清** — Type 收窄回「local 信息图的内部几何骨架」(11 个真骨架);原 4 个伪 type (hero/background/portrait/typography) 折回 `page_role: hero_page` + 4 条构图通则(single-subject / portrait / typographic / atmospheric);hero_page 文字分层规则(关键视觉词 embedded、可改文字走 SVG)-- **Brutalist AI 报章示例 deck 交付**([`examples/ppt169_brutalist_ai_newspaper_2026/`](../../examples/ppt169_brutalist_ai_newspaper_2026/)) — P0 三档第一档落地:满版小字 + 不规则栏宽 + halftone 黑白图 + 单点红 + 真原生 shape,10 页编辑部年报实压「文字位置精度 + 跨页一致性」-- **Kubernetes Blueprint 示例 deck 交付**([`examples/ppt169_kubernetes_blueprint_2026/`](../../examples/ppt169_kubernetes_blueprint_2026/)) — P0 三档第二档落地:等距工程图美学 + 蓝图青/琥珀色板 + 全手写 SVG 几何(无 raster 图)+ 自定义"逐笔绘制"动画,10 页 Kubernetes 架构走读实压「几何形状泛化 + chart 结构扩展性」-- **AI 图 `custom` 兜底出口** — `rendering` / `palette` / hero 构图三处允许声明 `custom` + 一段 `*_behavior` prose,替换原"找不到匹配就硬塞 vector-illustration / cool-corporate"的假兜底;端到端契约:[`image-renderings/_index.md`](../../skills/ppt-master/references/image-renderings/_index.md) §1.5 + [`image-palettes/_index.md`](../../skills/ppt-master/references/image-palettes/_index.md) §2 + [`strategist-image.md`](../../skills/ppt-master/references/strategist-image.md) §2 hard-rule(每维 ≤1 custom,单候选可双 custom)+ spec_lock 字段 + Image_Generator Step 2 消费分支-- **Template 架构三分类收口**([`docs/zh/templates-architecture.md`](./templates-architecture.md)) — brand / layout / deck 三独立目录 + 每类独立 schema + 段级合成 + git-style 冲突解决;[Generate PPTX Step 3](../../skills/ppt-master/workflows/generate-pptx.md#step-3-template-option) 按 kind 分支处理,触发规则仍是「显式路径才触发」-- **Pattern 填充 PPTX 安全网** — 被引用的 `<pattern>` 缺少 `data-pptx-pattern` 时,`svg_quality_checker.py` 现在给出 warning,转换器继续使用兼容的 `ltUpDiag` 回退;`patternTransform` 或超出 OOXML `ST_PresetPatternVal` 枚举的值仍然报错。[`native-data-interface.md` §1](../../skills/ppt-master/references/native-data-interface.md#1-pattern-fill--pattern-with-pptx-preset-annotation) 落地了完整 preset 清单,以及从导入元数据或子元素 paint 解析颜色的合同-- **LaTeX 数学公式渲染上线**([`scripts/latex_render.py`](../../skills/ppt-master/scripts/latex_render.py)) — Strategist 在 Typography 确认中锁定 `mixed` / `render-all` / `text-only` 三档策略,显式写 `images/formula_manifest.json`;脚本走 codecogs → quicklatex → mathpad → wikimedia 四源 fallback chain,输出透明 PNG 进 §VIII 表的 `Acquire Via: formula` / `Status: Rendered` 行;公式密集型 deck(学术 / 工程 / 教学)首次拥有原生渲染路径,规则面禁止扫源文件 `$...$` 自动渲染(公式选取是 Strategist 决策)-- **实时预览直接编辑 — L1 / L2 / L3**([`workflows/stages/live-preview.md`](../../skills/ppt-master/workflows/stages/live-preview.md)) — 浏览器编辑器新增无需 AI 往返的确定性就地编辑:文字内容(L1)、fill / stroke / font-size 等样式属性(L2)、以及画布上的几何操作(L3)——在选中元素上拖拽即移动、方向键微调(`Shift` = 10px)、多选、加右键重叠选择器选取堆叠元素。编辑支持 `Ctrl+Z` 撤销 + 合并,点 **Apply changes** 写回 `svg_output/`;移动经 finalize / 导出保位(移动的 text、提升的多行 tspan、重定位的 icon 都在 PPTX 中如实再现)。重新导出仍由对话触发;画布上的缩放手柄尚未实现(缩放走几何输入框)--### 2026-06(mode/视觉风格双 catalog + PPTX 入口与内容策略扩展)--- **复用原生 PPTX 设计 → 内容回填路线**([`workflows/template-fill-pptx.md`](../../skills/ppt-master/workflows/template-fill-pptx.md)) — 用户给一份现成 `.pptx` 加新材料 / 主题、要求「复用这套 deck 的设计 / 把内容填回去」时,Fill Native PPTX 路线直接编辑 PPTX,不进 SVG 生成管线。输出仍是原生可编辑 PPTX(复用原 slide 的形状 / 版式而非截图回填),过程做私有部件隔离、暴露图表数据、容量校验;触发同模板规则——显式要求复用既有 deck 才进,刻意不做改版式 / 加页 / 换图(那是 Generate PPTX 路线的职责)。与下文 Non-goals 的 #53 区分见该节-- **三个 executor 退役 → mode + visual-style 双 catalog**([`references/modes/`](../../skills/ppt-master/references/modes/) + [`references/visual-styles/`](../../skills/ppt-master/references/visual-styles/)) — 原三个 `executor-*.md`(general / consultant / consultant-top)把「领域 · 受众 · 说服 · 叙事」捆在一条线;拆成两个正交 catalog(照 `image-renderings` 范式:扁平目录 + `_index` + 按需读 + Strategist 锁一个)。**mode** = 讲解骨架(`pyramid` / `narrative` / `instructional` / `showcase`,consultant + top 因叙事内核相同合并为 pyramid);**visual-style** = SVG 排版美学(`swiss-minimal` / `editorial` / `soft-rounded` / `dark-tech`,各 paired 一个 image-rendering,**零 HEX**——颜色真值守在 confirmation e + image-palettes)。Strategist `§d` 双层独立锁定 `mode` + `visual_style` 进 `spec_lock`,Executor 加载两个 locked 文件;任意 mode × 任意 style 自由组合,渲染坐标仍留 `templates/charts/`-- **提示词约束强度三档解耦**([`docs/rules/prompt-style.md`](../rules/prompt-style.md) §4) — 规则(`Hard rule` / `Forbidden`)/ 默认(`Default — … may override`)/ 参考(`Reference — not a constraint`)三档显式化 + 「客观失败 vs 品味」判据 + checker 边界,让模型对「该守 vs 可破」一目了然;visual-style catalog 全程用 Reference 强度-- **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-structured.md`](../../skills/ppt-master/references/executor-structured.md) §1 套模板页重皮到当前 visual_style(模板供结构不供皮,mirror 模板保持已声明的视觉身份)、`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/stages/refine-spec.md`](../../skills/ppt-master/workflows/stages/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)) — 模板应用改由 AI 根据真实工作区和当前内容生成,确认页不再显示复用/遵循控件;strict/adaptive 仍作为 structured 导出器的内部值。确认页与 Live Preview 共用端口,最终 `result.json` 对用户可见选择权威。-- **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 profile 上线**([`workflows/profiles/beautify-pptx.md`](../../skills/ppt-master/workflows/profiles/beautify-pptx.md)) — 它是 Generate PPTX 的一个 profile,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,真要重排分页属普通 profile)、不保证坐标级 paste-back、combo / dual-axis / waterfall 图丢未捕获的绘图层--- **PPTX intake 多 deck 支持 + `analysis/` 源名前缀** — 主管线项目现在可把多份源 deck 合并进来:每份写 `<stem>.identity.json` / `<stem>.slide_library.json`,各自 digest 内联进单一索引 `source_profile.json` 的 `decks[]`(保住"Strategist 必读 `source_profile.json`"单入口契约,单 deck 即一条、多 deck 列多条;同 stem 重导覆盖该条)。`beautify` / `template-fill` 仍是 1:1 单 deck,按 stem 读自己那份 `<stem>.*`--- **材料发散度(Stage 1 的自由文字材料处理项)** — 主管线在使用情境 / 材料处理区加一个**纯文字**小问:用户用自己的话写要多贴源、还是多放开重塑(留空=平衡默认)。刻意不做固定档位、不按源信号替用户推荐、不联动页数——就是问用户本人意图。无论写得多放开都**事实守源**:只对源内内容重组 / 重框 / 展开 / 连结,绝不引入源外事实(那是 `topic-research` 阶段的职责)。Strategist 写 §IX 大纲时读这段 prose 消费、记 `design_spec §I`,**不进 spec_lock**(Executor 不读);`mode` 与发散度正交。beautify 会写入「原文与页序原样保留」并以锁定只读方式展示;template-fill 不进入这套确认流程,因此不展示此项--- **一批默认行为与入口标准化** — 逐元素入场动画默认关(只留转场 `fade`;元素动画改 opt-in `-a auto` / `animations.json`),消除"自动级联入场"的 AI 味;per-project `icons/` 在选择时把选中图标复制进项目、嵌入优先本地;`analysis/` 确立为机器抽取事实的 canonical 必读层(PPTX intake bundle + `image_analysis.csv`);主管线把源 deck 的身份(配色 / 字体 / 版式)当**参考而非约束**(可继承可重设,由策略师判断,默认从零设计);confirm 页支持自定义配色输入--- **AI 插画大图 → 切片点缀插画管线**([`scripts/slice_images.py`](../../skills/ppt-master/scripts/slice_images.py)) — 当一份 deck 需要若干同一家族的点缀插画时,不再一格一次 AI 调用,而是**一次生成一张多格插画大图**(单次调用锁住整组风格 / 色板、成本远低于逐格生成),再由 `slice_images.py` 确定性按 `RxC` 网格切成独立元素文件;`--trim` 按每格内容包围盒紧裁、`--alpha` 抠掉平整底色,让每个元素以**透明剪影**落到异色页面而非带可见方框。资源契约(§VIII)双行落地:一行 `ai` Illustration Sheet(生成但永不直接放置、不进 `spec_lock.md` images)+ 每格一行 `slice` 元素(实际放置、进 spec_lock);Step 5 生成大图后切片并重跑 `analyze_images`,Step 7 readiness gate 在离线场景列出大图 + 元素目标让用户手动放图再切。要不要用点缀插画是 Strategist 在 `image_usage` source 边界内的判断(不单独成确认字段;`image_usage: none` 永远压过插画意图),用户看不到内部 sheet/slice 实现。`svg_quality_checker` 加了对应校验,[`image-layout-patterns.md`](../../skills/ppt-master/references/image-layout-patterns.md) 补了图主导的促销 / 宣传版式范式--- **点缀插画:从「能切出来」到「会用、成体系」** — 在切片管线之上补齐决策层 + 收紧切片质量,让 deck 真正用好插画而非只是技术上能生成。①**切片质量**:`slice_images.py` 的 `--alpha` 改软蒙版抗锯齿 + 1px 去光晕、底色采样改 2px 边框环中位数、色距改最大通道差;`svg_quality_checker` 对 Generated `slice` 行校验文件存在性。②**触发倾向绑定 `visual_style`**:每个风格标 `core` / `supportive` / `sparse` 插画倾向([`visual-styles/_index.md`](../../skills/ppt-master/references/visual-styles/_index.md) 加 `Illus.` 列 + 各文件 §6),core 默认推荐用、sparse 默认不用;优先级链 `image_usage:none` → 用户显式意图(双向覆盖)→ 风格倾向 → none。③**贯穿母题 through-line**:deck 倾向用插画时,封面锚点 / 章节分隔 / 页内散点出自同一母题家族(共享 h.5 rendering+palette+主题世界),读成一套设计系统而非孤立散点;AI 母题仅在 `image_usage` 含 ai 时生成,provided/web 仅沿用本已成同族的素材。④**插画角色决策地图**:Strategist §h 加「角色 × 何时 × 机制 × source」导航表(散点 / 主角锚点 / 章节分隔 / 氛围背景 / 母题)。全程不设配额、不把品味数字化(同尺寸瓷砖检测评估后不做,留给 §4.3 placement 散文 + 执行判断)--- **图像变换矩阵端到端保真 + host-native 生成路径** — `svg_to_pptx` 的 DrawingML 图片导出现在如实尊重 SVG 的 transform 矩阵(旋转 / 斜切 / 复合变换不再在嵌套 `<g transform>` 下错位或塌回原点),把「跨四渲染器位置保真」主轴补到 raster 图层;`image_gen.py` 增加 host-native 生成路径,在宿主自带图像生成能力时走原生通道。两者均属修复 / 增量补强,细节见 commit log--- **网络配图改为「最佳图 + 可复核 + 人工换图」** — web 图来源不再默认静默下载一池候选,而是**默认只下最佳匹配图**,候选池退化成 `--save-candidates` 的显式升级路径(默认 4 张);每张下载图生成 ≤1024px review 副本(`images/.review/`,放置 / promote 仍全分辨率)。合适性复核做成 **model-agnostic**:多模态模型读 review 副本自查,非多模态则把 `source_page_url` 交人工判断——不假设模型有视觉。新增 `image_search.py --from-url <链接>`:把人找到的任意图片 URL 下载并替换目标(记 `license_tier: manual`、继承页面上下文),作为通用人工换图通道;`--promote` 改为从被选候选重算署名(不沿用旧图 credit)。全程在 Step 5 内、不合适转 `Needs-Manual` + 占位,**不阻塞主流程**。定位上 web 搜索是「兜底取图、不保证质量」,真要高质量靠 AI 生图或自己手动挑图换入--- **Web 配图实体安全门(精确主体不再被「高清错图」赢走)** — 承上条 web 配图:给 web 候选加 `required_terms` 实体门控,挡住「元数据相关但主体错误」的图(一张精修的罗马纪念碑赢下「重庆地标」行)。`required_terms` 各组之间 AND、组内 `A|B` 给别名(跨语言 `Chongqing|重庆`),匹配做小写 / 分隔符归一 / 空白压缩以兼容多词与 CJK 名;命中实体即视作相关信号(零 query 词重叠不再否决,CJK 标题地标可在英文 query 下通过,无 `required_terms` 时旧的否决逻辑照旧)。像素面积从主导分(cap 5000)降级为 tie-breaker(cap 1500)+ 标题命中加权,让实体准确性与相关性压过纯分辨率(避免高清错主体赢)。CLI `--require-terms`(可重复,逗号 / `|`)、批量 `required_terms`、`--from-url` 均继承,记入 `image_sources.json` 备审;门控形同虚设(弱 `required_terms`)会发 warning。定位是与 `.review` 视觉复核配对的**元数据门**,不是视觉分类器--- **原生 PPTX 导出图片媒体大小封顶** — 保持生成 deck 可编辑、不嵌入巨幅源图:新增原生图片尺寸模式——`cap`(默认)只对超大源图限制最大边长,`display` 按渲染 SVG 框尺寸做更激进压缩;原生导出保留完整嵌入像素,SVG/PPT 显示裁剪仍走可编辑的 picture-crop 元数据;`finalize_svg` 保留原有 slice/meet 行为,另加默认按渲染尺寸下采样以产出紧凑 SVG 快照。文档落地 `cap` / `display` 两模式与 `--no-image-optimize` 逃生舱--### 2026-07(分阶段确认 UI + 原生对象 + 动效加固)--- **Step 4 确认 gate 重构为三阶段向导 + 可视化预览** — 原来单次「八项确认」gate 拆成一个浏览器会话内按依赖排序的流程:开放式沟通契约 → 完整 PPT 方案 → 资源与生产执行。沟通目的用自然语言保留多个目标及其关系;常见目的只作提示,不设 `primary_job` 点选项。Stage 1 每个可编辑文本框里都是推荐,不是必填答案;确认时按当前值原样落盘,用户清空的字段会一直保持为空,直到最终结果。Stage 2 用稳妥 / 偏移 / 大胆三套成套方向协调视觉风格、色彩、字体、图标与生成图渲染;Stage 3 不再出现审美选择。生成图只确认渲染风格,颜色直接继承已选的整套 PPT 配色。每个下游阶段都从用户**实际已确认**的上游答案推导。`recommendations.json` 使用规范的 `stage` 选择器(`tier` 只作只读兼容输入);聊天 fallback 镜像同样的分阶段顺序-- **取消独立的 AI 图调色确认** — 五月建立的 rendering × palette × type 结构继续兼容旧项目和历史对比资产,但新推荐与新锁不再写 `image_palette`。Stage 2 在每套完整设计方向里只确认生成图渲染;Image_Generator 直接消费已选 PPT 的色彩角色,从根上去掉重复色彩决策和同阶段 palette / HEX 失配--- **页间转场与元素入场动画完成无静默降级加固** — 当前默认保持页间 `fade` / 0.4 秒、元素入场 `none`,对象动画仍通过 `-a` 或 `animations.json` 按需开启。未知效果 / Start 模式、非有限或越界时长、非法顺序,以及缺失 slide/group 引用都会直接失败,不再偷换为 `fade`、其它 Start 模式或继承值。公开产物替换前会回读候选 PPTX,校验根级 timing 位置、时间节点 ID 唯一性、shape 引用、效果 / 时长 / Start 语义及旁白 timing 合并。直接 PPTX 路线只保留源对象动画,不把它翻译成生成路线的动画模型。Microsoft PowerPoint 是动效行为的主要验证目标;其它演示软件仅作为兼容目标,不承诺完全相同的播放结果+----- **`--native-charts-and-tables` 从休眠 marker 硬化为可用级 opt-in** — 那条窄「原生 Chart/Table」例外(见下文 Non-goals)现在导出的图表与纯文本表格会**保留 deck 自己的设计**,不再塌回 PowerPoint 的白底默认主题。classic 原生图表显式写入 chart-area / plot-area / 轴线 / 网格线 / 标签文字颜色——从可见的 SVG fallback 推断(最大面板型 `<rect>` → 背景、fallback 文字 → 标签、fallback 描边 → 轴线/网格),或用 `style` 显式覆盖(`chart_area_fill` / `plot_area_fill` / `text_color` / `axis_color` / `grid_color`,`"none"` 表透明);颜色解析把命名色、`#RGB` 简写、`rgb()` / `rgba()` 归一为 OOXML hex;bar/column 系列关掉负值反色,负值柱保持系列色。激活导出命名为 `<name>_<ts>_native_charts_tables.pptx`,以与默认形状导出区分。**默认路线不变**——图表/表格仍以可编辑的 SVG 派生 DrawingML 形状导出以保跨渲染器保真;PowerPoint 原生 Chart/Table 增加数据源和对象专属编辑模型,仍是下文 Non-goals 里那条刻意的 opt-in 取舍+## 进行中 / 下一步-- **原生 package 结构 + 按模式创建模板** — deck/layout-template SVG 页面在创作时就声明最终 Master/Layout 身份。固定 Master/Layout 视觉是根级原子,可复用槽位是带真实 carrier 或显式 composite proxy 的有界顶层 group,零槽位 Layout 也合法。`structured` 导出只确定性编译该合同并执行最终 package 回读;不提升重复 chrome,也不推断 placeholder。`flat` 自由设计 / brand-only 导出则保持所有内容 Slide-local,同时把 stock Office 脚手架替换成一个属于项目的干净 Master、一个 Blank Layout 和按 deck 命名的当前 lock 主题;删除 title/body 等 stock 内容、裁掉未使用 Layout,仅保留标准日期、页脚和页码能力钩子。两种生成模式都把锁定 title 与确定性的九级 body 层级写入母版 `p:txStyles`,同时保留段落和项目符号设置。`standard` / `fidelity` 重新创作 SVG roster 和新的 Master/Layout 系统,不保留、也不蒸馏来源拓扑。`mirror` 从原生来源包内实际存在的全部受支持事实物化新工作区,包括未使用 Layout 的定义专用原型;固定结构层的来源 group 只允许机械展开成直接原子,不做语义归纳。无损导入在分析区保持不可变并作为载荷后备,`authoring-svg/` 及其 manifest 则组成可编辑的模板创建 IR。`library` 与 `project` 都要求 `templates/`,`images/` / `icons/` 可选,`exports/` 仅在按需生成评审文件时出现;只有全局注册不同。旧 baseline/template/preserve SVG 包不做迁移:只把它们作为视觉参考,由 `create-template` 创作新工作区,再从新工作区生成新 PPTX;原始 PPTX 的一次性回填仍走 `template-fill-pptx`。+明确在做或下一步要做,不承诺时间窗口。-- **原生预设创作引导跨角色强化** — 用真正的 PowerPoint 预设(`prstGeom`,带调节手柄、非扁平卡片外观)被重申为库存几何(块箭头、chevron、横幅、标注、流程节点、星形)的**默认**而非例外。新页面与项目自有模板使用 `preset_shape_svg.py` 输出的 compact 原子 `<g>`:group 只写一次 preset metadata 与基础 paint,直接可见 path 是 registry 派生层,只保留必要的分层 paint 覆盖。创作形式没有 hidden carrier、preview wrapper 或持久化 fingerprint;只允许通过 helper 生成,preset、frame、adjustment 或 paint 任一变化都整体重生成,不手改 path 或 metadata。因此,当 compact 片段明确表达所需库存形状时,可以保存在 `templates/charts/` 中。Executor 仍按对象意图在画图当下决定,绝不扫描已完成路径;paint 边界保持收窄:渐变填充/描边或 pattern 填充仍用普通 SVG。PPTX 导入与 `mirror` 继续使用另一套 expanded 无损表达。Strategist 仍给可能适用的 §VII 页面追加非破坏性的 `native-preset candidate` 注记;具体 preset、frame 与 paint 由 Executor 选择+- **在真实 deck 上校准新落地的体系** — 多 deck 合并 intake、材料发散度、插画体系、结构化模板创作均已上线;它们现在需要的是真实使用信号,而不是更多机制。不预先加机械阈值或配额。+- **Prompt 精简** — 在不降质量的前提下压缩各角色 prompt 的 token 占用、提升缓存命中率,带来间接的成本 / 速度改善。与「纯速度优化」的边界见下方「明确不做」。----## 进行中 / 下一步+## 未来方向(信号驱动)-明确在做或下一步要做的方向,不承诺时间窗口。+已评估为「真实需求出现时值得做」的候选项,列出来是为了公开意图,均不构成承诺。-- **多 deck intake 与材料发散度的真实使用校准(刚落地)** — 多 deck 合并 intake(`<stem>` 前缀 + `decks[]` 合并索引)与 Stage 1 的材料发散度自由文字项均已上线(见上「2026-06」),接下来按真实使用信号校准:多份源 deck 同名(stem 冲突)的处理目前是后者覆盖前者,是否需要去重 / 加序号待信号;发散度的自由文字让 Strategist 判得准不准、放开写时「事实守源」边界守不守得住,待真实生成验证。两者都不预先加机械阈值-- **插画能力(机制 + 部署层)的真实 deck 校准(刚落地)** — 切片管线、边缘质量收紧,以及决策层(风格倾向 / 贯穿母题 / 角色地图,见上「2026-06」)均已上线,接下来按真实使用信号校准:一次大图切多格的风格 / 色板一致性与 `--alpha` 软蒙版对格内不规则构图的鲁棒性、离线 readiness gate 的手动放图 + 重切体验、风格倾向是否翻对了该翻的风格、母题在真实 deck 上读成「设计系统」还是「过度装饰」、以及 source 边界(provided/web 不静默生成 AI)守得住否。不预先加机械阈值 / 配额;同尺寸瓷砖若真反复出现再考虑更窄的 lint-- **创作预设上的原生投影——作为独立增强延后** — compact authored-preset `<g>` 是封闭的语义原子,当前只允许已登记的 preset metadata、基础 paint 和直接 registry path;`filter` / effect metadata 不在创作合同内。转换器可能在内部把该原子展开为既有 carrier/preview 传输结构,但那是实现细节,不是源 SVG 接口。PPTX 导入与 `mirror` 可以在另一套 expanded 无损表达中保留受支持的来源效果,但这不会放宽新创作合同。在形成精确的 preset-effect 合同并补齐 checker 覆盖前,需要阴影的库存形状保守留普通 SVG;是否扩展继续以真实需求为依据,不把它表述成现有能力-- 其余:mode / visual-style 体系的验证与校准已收口(见上「2026-06」),结构(5 mode + 18 visual-style + custom)定型、四对近邻消歧并成一张 Close-calls 表、四项校准收紧已落地。后续方向由真实使用信号与反馈驱动;长期改进见下「持续维护方向」,已评估不做的见「明确不做」+- **持续收窄[映射指南](./powerpoint-svg-mapping.md)记录的原生覆盖缺口** — 逐版本把更多「仅 SVG」的格子推向 PowerPoint 原生结构与行为。+- **创作型预设形状的效果支持**(如原生阴影)— 等形成精确的 preset-effect 契约并补齐 checker 覆盖再做;在此之前,需要阴影的库存形状保守留普通 SVG。+- **生成侧超链接创作** — 源 deck 里已有的超链接如今能在转换中保留;让 Strategist 主动创作新链接,等需求出现再做。+- **图片页面背景提升为原生背景填充** — 纯色 / 渐变页面背景已导出为 PowerPoint 原生底色;图片背景按需求驱动。----## 持续维护方向+## 已交付里程碑-不承诺时间窗口的长期改进项。只列真方向,具体修复 / 单 flag 看 commit log。+一个月一行,细节见 [Release 发布说明](https://github.com/hugohe3/ppt-master/releases)与 commit log。-- **Prompt 精简** — 在不降质量的前提下压缩各角色 prompt 的 token 占用、提升缓存命中率,带来间接的成本 / 速度改善。与下面「纯速度优化」一节互补:做间接优化,不做牺牲质量的提速。+| 时间 | 主题 |+|---|---|+| 2026-03 | **原生 PPTX 路线成形** — SVG → DrawingML 链路可用;图表 / 版式模板索引上线 |+| 2026-04 | **管线规模化** — 仅凭主题生成、70 个图表模板 + 三套图标库、`spec_lock` 跨页一致性契约、逐元素动画与旁白 / 视频导出 |+| 2026-05 | **可视化编辑 + AI 图片体系化** — Live Preview 确定性原位编辑(基于 [@WodenJay](https://github.com/WodenJay) 的 [PR #85](https://github.com/hugohe3/ppt-master/pull/85))、从 PPTX 创建模板工作区、rendering × palette × type 图片体系、LaTeX 公式渲染 |+| 2026-06 | **mode 与 visual-style 双 catalog + intake 扩展** — 5 种叙事 mode × 18 种视觉风格(+ `custom`)、内容忠实的美化 profile、多 deck 合并 intake、插画切片管线、网络图片质量闸门、源转换保真提升(图注识别基于 [@suay1113](https://github.com/suay1113) 的 [PR #191](https://github.com/hugohe3/ppt-master/pull/191),超链接保留提炼自 [@ZhaoZuohong](https://github.com/ZhaoZuohong) 的 [PR #155](https://github.com/hugohe3/ppt-master/pull/155)) |+| 2026-07 | **定位章程 + 原生母版 / 版式 + token 效率**([v4.0.0](https://github.com/hugohe3/ppt-master/releases/tag/v4.0.0))— 三段式分步确认 UI、真 `p:sldMaster` / `p:sldLayout` 导出、`--native-charts-and-tables` opt-in、动效导出加固、图表模板库压缩 |---@@ -130,7 +70,7 @@ Generate PPTX 路线围绕完全可控的新形状、文字与版式创作。结跨四渲染器(PowerPoint / Keynote / LibreOffice / WPS)的位置保真是项目主轴。把默认路线改成 PowerPoint 原生图表会让「像素级一致性」破功——同一个 PPTX 在不同渲染器里图表会显示不同布局。图表默认用 SVG 是 **by design**,不是能力缺失。-窄例外是 `data-pptx-replace-with` marker:受支持的数据图表与纯文本网格表格在生成时携带 PowerPoint 原生 Chart/Table 替换 payload,导出加 `--native-charts-and-tables` 才激活——供主动用跨渲染器保真换取带数据源对象及图表/表格专属编辑模型的用户使用;激活后的对象会保留 deck 的 chart-area / plot / 轴线 / 网格线 / 标签颜色与原生表格格式,不再塌回 PowerPoint 默认主题(见上文 2026-07)。默认导出路径与可编辑的 SVG 派生形状系统不变。+窄例外是 `data-pptx-replace-with` marker:受支持的数据图表与纯文本网格表格在生成时携带 PowerPoint 原生 Chart/Table 替换 payload,导出加 `--native-charts-and-tables` 才激活——供主动用跨渲染器保真换取带数据源对象及图表/表格专属编辑模型的用户使用;激活后的对象会保留 deck 的 chart-area / plot / 轴线 / 网格线 / 标签颜色与原生表格格式,不再塌回 PowerPoint 默认主题(见 [v4.0.0 发布说明](https://github.com/hugohe3/ppt-master/releases/tag/v4.0.0))。默认导出路径与可编辑的 SVG 派生形状系统不变。### uv 作为默认 / 必需依赖