docs: add remotion workflow rules

Add deck project text records
2026-07-07 18:11:45 +08:00 · 2026-07-07 17:11:01 +08:00
184 changed files with 13618 additions and 42 deletions
--- a/.gitignore
+++ b/.gitignore
@ -17,7 +17,23 @@ projects/**/slides/s*/*.mkv
 projects/**/slides/s*/*.wav
 projects/**/slides/s*/*.mp3
 projects/**/slides/s*/*.m4a
+projects/**/slides/*.pdf
+projects/**/slides/*.pptx
 projects/**/renders/**
 projects/**/exports/**
 projects/**/videos/**
 projects/**/archive/**
+
+# Remotion / Node project generated files
+projects/**/remotion/node_modules/
+projects/**/remotion/.cache/
+projects/**/remotion/.turbo/
+projects/**/remotion/out/
+projects/**/remotion/dist/
+projects/**/remotion/build/
+projects/**/remotion/.next/
+projects/**/remotion/*.mp4
+projects/**/remotion/*.mov
+projects/**/remotion/*.webm
+projects/**/remotion/*.avi
+projects/**/remotion/*.mkv
--- a/AGENTS.md
+++ b/AGENTS.md
@ -84,9 +84,42 @@ For every project-bound image prompt document:

 Project-bound image generation must remain approval-gated: save the prompt/spec first, wait for user approval, then execute generation only when explicitly requested. A task-level user instruction that records the running mode and slide scope counts as explicit generation approval for that task scope. If the task mode is `prompt only`, do not generate images.

+## Remotion Workflow
+
+Use this workflow when accepted planning material, the user, or the active task says the video should be implemented in Remotion, or when Codex writes or edits Remotion code.
+
+Remotion planning from `强哥的策划导演3.0` is a planning source, not an execution package. GPT may provide Remotion adaptation judgments, a Remotion style-system anchor, storyboard-to-component mapping, and a Remotion Implementation Brief. GPT must not be treated as the source for final code paths, dependency commands, render commands, or complete Remotion source code.
+
+Default local source root:
+
+```text
+projects/<project>/remotion/
+```
+
+Use this path for one-off project-specific Remotion source. Reusable products, reusable templates, deployable systems, or business software belong in the relevant `work-projects` repository instead of Video Workbench unless the user explicitly keeps them local for a prototype.
+
+Default Remotion deliverable:
+
+```text
+source code + renderable MP4
+```
+
+For project-bound Remotion work:
+
+- use worker profile `remotion-video-production`;
+- read `C:\Users\wangq\.agents\skills\remotion\SKILL.md` before writing or editing Remotion code;
+- before visual scenes, layouts, motion graphics, or text-heavy Remotion screens, also read the Remotion Skill's relevant rule files, starting with `rules/video-layout.md`;
+- load narrower Remotion Skill rule files when the task involves captions, audio, transitions, charts, maps, 3D, effects, timing, images, fonts, or parameterized videos;
+- keep Remotion assets referenced by source code under `projects/<project>/remotion/public/` unless the project has a better local convention;
+- place rendered MP4s or render tests under `projects/<project>/renders/`, `projects/<project>/exports/`, or another path recorded in `execution-plan.md`;
+- keep generated render outputs out of Git by default; record their paths, status, and hashes when useful for continuation;
+- run narrow code/render checks for Remotion work, such as typecheck/build when available and `npx remotion still` or an explicit render command for the target composition. These checks are part of code output validation, not optional image review.
+
+The user has given standing task-level permission for first and future Remotion tests to use `npx create-video`, npm dependency installation, and Remotion render commands when the active task needs them. Tooling may still require sandbox/network escalation approval at command time.
+
 ## Slide Task Workflow

-For long slide / video production work, use the local slide task orchestration workflow:
+For long slide / video production work, including Remotion scene or section production, use the local slide task orchestration workflow:

 ```text
 docs/workflows/slide-task-orchestration.md
@ -147,6 +180,7 @@ Rules:
 - Production-light child sessions write per-slide briefs, prompts, images, and compact handoff directly; they may handle at most 8 ordinary production slides by default. Reduce to 1-3 slides for identity-heavy, repair-heavy, core-risk, or context-expensive scopes.
 - Test/audit child sessions may launch slide agents; a slide agent handles exactly 1 slide per invocation.
 - The default worker profile for current MV / video work is `video-slide-production`.
+- Use `remotion-video-production` when the accepted plan calls for Remotion implementation, Remotion scene/component production, or a renderable MP4 from React/TypeScript source.
 - Reserved future profile `ppt-page-production` may use `docs/cases/`, but must be designed when the first real PPT task needs it.
 - For cover pages, PPT pages, science-video pages, title cards, explainer pages, or any slide that needs later human layout / typography / post-production, add a separate `sNN-vN-design.md` before `brief`. `design` is for the user and records visual concept, layout, text content, typography/color guidance, post-production notes, constraints, and references. `brief` remains the execution card for Codex / image generation and must not absorb human-facing design or post-production notes.
 - `sNN-vN-design.md` is Chinese by default. If an English design companion is needed, write `sNN-vN-design-en.md` beside it; the English file supplements but does not replace the default Chinese design.
@ -157,7 +191,10 @@ Rules:
 - Character continuity must use `visual-system/characters/character-reference-registry.md` and active character image references.
 - Visible character identity cannot be generated from text-only prompts when identity matters.
 - Slide prompts must use the `gpt-image-2` Skill Advisor / host-native workflow.
+- Remotion code must use the installed Remotion Skill workflow and should not use `gpt-image-2` unless the Remotion project separately needs generated still assets.
 - In `slides/sNN/`, keep Markdown control/script artifacts in Git, but treat generated images, audio, and video as asset-layer files that are ignored by default and recorded by path/hash/status in control files.
+- In `projects/<project>/remotion/`, keep source code, package files, and source assets in Git by default; keep `node_modules`, caches, build outputs, and rendered media ignored by default.
+- Assembled `.pptx` / `.pdf` files under project `slides/` are generated deliverables. Do not commit them unless the user explicitly promotes a named file as source-authority material.
 - Project `archive/` directories are for discarded images and discarded documents by default. Do not add them to Git unless the user explicitly promotes a specific archived item back into `intake/`, `visual-system/`, or another source-authority location.
 - New slide-task execution must append the next unused `sNN-vN-*` version and must not overwrite existing slide artifacts unless the user explicitly authorizes an overwrite repair for named files.
 - Review defaults to `none` in production-light and local-only in test/audit or explicit review tasks. Do not call Minimax MCP or other external vision/review services unless the user explicitly authorizes that task.
--- a/PROJECTS.md
+++ b/PROJECTS.md
@ -15,6 +15,7 @@ It turns accepted planning material and deep source references into public-facin
 - video outlines;
 - storyboard logic;
 - shot-level production files;
+- project-specific Remotion source and render receipts;
 - voiceover scripts;
 - image prompts;
 - asset manifests;
@ -31,7 +32,7 @@ It does not own upstream deep creation, canonical agents, reusable automation so
 | `writing-workbench` | Sibling workspace for deep article production | completed or sufficiently stable article projects when the user chooses to make video content | no default output |
 | `ccpe-system` | Supplier of local work agents, runtime rules, model cards, and governed cognitive assets | the external `Interface.md`, the consumer setup contract, and task-specific canonical artifacts only when routed by the setup/registry | capability requests under `requirements/ccpe/` |
 | `skills-vault` | Supplier of deterministic automation tools | file processors, batch utilities, asset tools, export tools, installable automation skills | capability requests under `requirements/skills-vault/` |
-| `work-projects` | Separate software/customer project workspace | no default input | production system requirements if this video workflow is later packaged as deployable software |
+| `work-projects` | Separate software/customer project workspace | no default input | production system requirements if this video workflow is later packaged as deployable software or a reusable Remotion product/template |

 ## Boundary Rules

@ -40,6 +41,7 @@ It does not own upstream deep creation, canonical agents, reusable automation so
 - Long-term sources, finished essays, model documents, and open-ended discussion memory belong in `knowledge-vault`.
 - Canonical local-work agents, prompts, runtimes, model cards, and invocation protocols belong in `ccpe-system`; Video Workbench starts from `ccpe-system\Interface.md`, not from hardcoded internal CCPE artifact paths.
 - Deterministic reusable automation implementation source belongs in `skills-vault`.
+- One-off project-specific Remotion source belongs under `projects/<project>/remotion/`; reusable Remotion products, reusable templates, deployable systems, and customer/business software belong in `work-projects`.
 - If the video workflow is later packaged as a deployable system, production/business agent implementation belongs in the relevant `work-projects` repository, not in this local workbench.
 - Missing capabilities are requested through `requirements/`, not implemented ad hoc inside this repository.

--- a/README.md
+++ b/README.md
@ -1,15 +1,15 @@
 # Video Workbench

-Video Workbench is the local execution workspace for dimensional video and slide-style output. It receives GPT V2 planning Markdown from the user, keeps lightweight project records, and produces local execution plans, visual-system assets, prompts, images, narration audio, slide/page files, and handoff notes.
+Video Workbench is the local execution workspace for dimensional video, Remotion, and slide-style output. It receives GPT planning Markdown from the user, keeps lightweight project records, and produces local execution plans, visual-system assets, prompts, images, Remotion source projects, narration audio, renderable video files, slide/page files, and handoff notes.

 It is a sibling of Writing Workbench, not a subfolder of it.

 ## Working Model

 ```text
-GPT / ChatGPT      = stage 0-5 planning Markdown, outlines, visual systems, page/shot skeletons
+GPT / ChatGPT      = stage 0-5 planning Markdown, outlines, visual systems, page/shot skeletons, Remotion planning briefs
 User               = decisions, source material, accepted planning files
-Video Workbench    = stage 6+ execution planning, Prompt Advisor work, visual-system materialization, generated assets, handoff
+Video Workbench    = stage 6+ execution planning, Prompt Advisor work, Remotion implementation, visual-system materialization, generated assets, handoff
 CCPE System        = local work agents, runtimes, model cards, interfaces
 Skills Vault       = reusable automation skills and deterministic tools
 ```
@ -28,6 +28,10 @@ projects/
      visual-system.md
      refs/
      iterations/
+    remotion/
+      package.json
+      src/
+      public/
    slides/
      slides.md
      s01/
@ -43,10 +47,11 @@ ccpe-consumption/
 ```

 - `projects/` is the only project entry point. Project status and series membership are recorded inside project files, not by moving directories.
- `projects/<project>/intake/` holds user-provided GPT V2 stage 0-5 Markdown, source notes, references, and other accepted inputs.
+- `projects/<project>/intake/` holds user-provided GPT stage 0-5 Markdown, source notes, references, and other accepted inputs.
 - `projects/<project>/project.md` is a Codex-maintained project note and file map. It should link to content in `intake/`, not duplicate it.
 - `projects/<project>/execution-plan.md` is the project-level execution control file for visual work. It records the current strategy, batch, round, and next decision instead of duplicating the project map.
 - `projects/<project>/visual-system/visual-system.md` records the current accepted visual system, materialization targets, iteration log, failure attribution, and next action. Reference images and iteration artifacts stay under `refs/` and `iterations/`.
+- `projects/<project>/remotion/` is the default source root for one-off project-specific Remotion videos. Keep source code, package files, and source assets there; keep generated renders, caches, and dependencies out of Git by default.
 - `projects/<project>/slides/` is the working area for slide, page, or shot execution. Each `sNN/` folder keeps that unit's prompts, narration, generated images, audio, and local notes together. Files inside `sNN/` must carry the `sNN` prefix, and iteration files use `sNN-vN-type.ext`, such as `s03-v1-prompt.md`.
 - `handoff/` is for repo-level, cross-project, or context-window handoff notes.
 - `docs/cases/` is the Codex-side local case-pattern library extracted from old real GPT cases. It preserves execution patterns, not old prompts, JSON packages, paths, or global text-rendering rules.
@ -58,8 +63,10 @@ Series indexes can be added later under `projects/_series/` when a real series n
 ## Boundaries

 - Manual editing and publishing remain human-controlled unless explicitly automated later.
- Video Workbench does not default to full video editing, image-to-video generation, or platform publishing.
- Codex does not ask GPT V2 for final image prompts, Codex JSON, output paths, task lists, or image generation parameters.
+- Video Workbench does not default to full manual video editing, image-to-video generation, or platform publishing.
+- Codex does not ask GPT for final image prompts, Codex JSON, output paths, task lists, dependency commands, render commands, or complete Remotion source code.
+- For Remotion videos, accepted GPT planning can supply adaptation judgment, a style-system anchor, storyboard-to-component mapping, and a Remotion Implementation Brief. Codex owns the local source layout, implementation, dependency commands, render commands, and validation.
+- The default Remotion deliverable is source code plus a renderable MP4. The user has standing task-level permission for `npx create-video`, npm dependency installation, and Remotion render commands when a Remotion task needs them; sandbox or network approvals may still be requested at command time.
 - `makeup-still/` is not a recommended project directory. Character, scene, template, diagram, and background anchors belong under `visual-system/` when needed.
 - `tmp/` is deprecated. New reusable or reviewable investigation material belongs under `investigations/`.
 - Missing CCPE assets should be requested under `requirements/ccpe/`.
--- a/VIDEO_WORKBENCH.md
+++ b/VIDEO_WORKBENCH.md
@ -2,19 +2,21 @@

 ## 1. Workspace Identity

-This project is the local execution workspace for dimensional video and slide-style output.
+This project is the local execution workspace for dimensional video, Remotion, and slide-style output.

-It turns accepted GPT V2 planning Markdown and user-provided sources into project notes, local execution plans, visual-system assets, slide/page execution folders, generated assets, and compact handoff material. It is optimized for practical continuation and production throughput, not heavy governance.
+It turns accepted GPT planning Markdown and user-provided sources into project notes, local execution plans, visual-system assets, slide/page execution folders, Remotion source projects, generated assets, renderable video files, and compact handoff material. It is optimized for practical continuation and production throughput, not heavy governance.

 ## 2. Collaboration Roles

 ```text
-GPT / ChatGPT   = stage 0-5 planning Markdown, outlines, visual systems, page/shot skeletons
+GPT / ChatGPT   = stage 0-5 planning Markdown, outlines, visual systems, page/shot skeletons, Remotion planning briefs
 User            = decisions, accepted inputs, source/reference material
-Codex           = stage 6+ execution planning, Prompt Advisor work, visual-system materialization, iteration, local assets
+Codex           = stage 6+ execution planning, Prompt Advisor work, Remotion implementation, visual-system materialization, iteration, local assets
 ```

-GPT V2 does not provide final image prompts, Codex JSON, output paths, task lists, or image generation parameters. Codex should not request those from GPT as missing inputs.
+GPT does not provide final image prompts, Codex JSON, output paths, task lists, dependency commands, render commands, complete Remotion source code, or image generation parameters. Codex should not request those from GPT as missing inputs.
+
+For Remotion videos, accepted planning from `强哥的策划导演3.0` may provide Remotion adaptation judgment, a Remotion style-system anchor, storyboard-to-component mapping, and a Remotion Implementation Brief. The brief is a planning source, not a Codex execution package.

 Codex should not rewrite GPT planning into duplicate project content. The accepted planning files belong in the project `intake/` directory, and Codex turns them into local execution decisions in `execution-plan.md`.

@ -32,6 +34,10 @@ projects/
      visual-system.md
      refs/
      iterations/
+    remotion/
+      package.json
+      src/
+      public/
    slides/
      slides.md
      s01/
@ -55,8 +61,10 @@ If a series needs more structure, create `projects/_series/` when the first real
 - lifecycle phase from `docs/workflows/project-lifecycle.md`;
 - workflow phase: new projects default to `test/audit`; switch to `production-light` only after explicit user authorization;
 - medium type;
+- worker profile, such as `video-slide-production`, `remotion-video-production`, or a future profile;
 - source and reference links;
- which `intake/` files are GPT V2 planning Markdown, source notes, reference notes, or background notes;
+- which `intake/` files are GPT planning Markdown, source notes, reference notes, or background notes;
+- Remotion source root when applicable, defaulting to `projects/<project>/remotion/`;
 - current execution focus;
 - key local execution files;
 - next action.
@ -81,8 +89,10 @@ This is a living Codex-maintained execution control file. It records:

 - the current medium branch;
 - the current lifecycle phase when it affects execution;
+- the active worker profile;
 - the current execution focus;
 - which visual-system assets need materialization;
+- for Remotion projects, source root, package manager state, composition IDs, render target, render output path, and validation commands;
 - the current small-batch iteration strategy;
 - which `sNN` units are in the active batch;
 - the current round;
@ -94,9 +104,10 @@ It should not repeat the project ID, full intake file index, or project map from

 `projects/<project>/intake/` holds material provided or accepted by the user:

- GPT V2 stage 0-5 planning Markdown;
+- GPT stage 0-5 planning Markdown;
 - outlines;
 - storyboards, page skeletons, or shot skeletons;
+- Remotion adaptation judgments, Remotion style-system anchors, Remotion storyboards, storyboard-to-component mappings, and Remotion Implementation Briefs;
 - background notes;
 - reference images, audio notes, or source pointers.

@ -171,10 +182,13 @@ visual-system/scenes/
 visual-system/templates/
 visual-system/diagrams/
 visual-system/backgrounds/
+visual-system/remotion/
 ```

 Do not create a separate `makeup-still/` directory. Character anchors and identity references belong under `visual-system/`.

+For Remotion projects, record the accepted Remotion style-system anchor under `visual-system/remotion/` or in `visual-system/visual-system.md` if the project is small. It should cover background systems, color and type hierarchy, component style, information density, animation language, transition motifs, chart/flow/formula handling, subtitle strategy, and reusable component suggestions. This is source authority for the Remotion implementation; it is not React/TypeScript code.
+
 After a character anchor stage is accepted, clean the character workspace before scene generation:

 - keep only current usable character references under `visual-system/characters/`;
@ -186,11 +200,11 @@ For identity-critical shots, prompt specs must choose the person reference from

 ## 8. Slides And Shots

-Use `slides/` as the execution unit area for both video shots and PPT-style pages.
+Use `slides/` as the execution unit area for video shots, PPT-style pages, Remotion scenes, and Remotion sections.

-`projects/<project>/slides/slides.md` is the unit execution fact table. It records each page or shot's source, batch, current prompt version, current image version, review state, acceptance state, and notes. Project-level iteration strategy belongs in `execution-plan.md`.
+`projects/<project>/slides/slides.md` is the unit execution fact table. It records each page, shot, Remotion scene, or Remotion section's source, batch, current prompt/spec version, current image/render/component status, review state, acceptance state, and notes. Project-level iteration strategy belongs in `execution-plan.md`.

-Each `slides/sNN/` folder keeps one slide, page, or shot together:
+Each `slides/sNN/` folder keeps one slide, page, shot, Remotion scene, or Remotion section together:

 ```text
 slides/
@ -203,6 +217,8 @@ slides/
    s01-visual-brief.md
    s01-v1-prompt.md
    s01-v1-image.png
+    s01-v1-remotion-spec.md
+    s01-v1-render-check.md
    s01-v1-review.md      # optional; test/audit or explicit review only
    s01-v2-prompt.md
    s01-v2-image.png
@ -211,11 +227,13 @@ slides/

 All files inside `sNN/` must carry the `sNN` prefix. Iteration files use `sNN-vN-type.ext`, so files from the same version sort together.

-`design` is a human-facing design specification, not an image-generation execution card. Use `sNN-vN-design.md` for cover pages, PPT pages, science-video pages, title cards, explainers, and any slide that needs later human layout, typography, text placement, or post-production. It records the design concept, intended text, placement, typography/color guidance, constraints, references, and post-production notes. `brief` stays compact and execution-facing for Codex / image generation.
+`design` is a human-facing design specification, not an image-generation execution card. Use `sNN-vN-design.md` for cover pages, PPT pages, science-video pages, title cards, explainers, Remotion scenes with meaningful typography/layout choices, and any slide that needs later human layout, typography, text placement, or post-production. It records the design concept, intended text, placement, typography/color guidance, constraints, references, and post-production notes. `brief` stays compact and execution-facing for Codex / image generation or Remotion implementation.

 `sNN-vN-design.md` is Chinese by default. If an English design companion is needed, write `sNN-vN-design-en.md` beside it. The English companion is optional and never replaces the Chinese default.

-For design-gated slides, write `design` first and wait for user confirmation before writing `brief`, `prompt`, or generating images, unless the user explicitly authorizes the whole chain. Ordinary MV shot production may skip `design` and continue directly with `brief -> prompt -> image`.
+For design-gated slides, write `design` first and wait for user confirmation before writing `brief`, `prompt`, Remotion specs, or generating/rendering outputs, unless the user explicitly authorizes the whole chain. Ordinary MV shot production may skip `design` and continue directly with `brief -> prompt -> image`.
+
+For Remotion work, `slides/sNN/` holds scene/section intent and checks. React/TypeScript source code lives under `projects/<project>/remotion/`. Do not store normal component source as append-only `sNN-vN-*` Markdown; use Git history for code changes and use `slides/sNN/` for the scene spec, component mapping, and render evidence.

 New task execution must append a new slide version instead of overwriting an existing one. Resolve the next unused `vN` from existing `sNN-v*-*` files in the slide folder, and treat incomplete or superseded earlier versions as historical evidence rather than paths to reuse.

@ -247,17 +265,18 @@ This file records the target scope, completed units, active child session, next
 Roles:

 - Main session: receives user instructions, defines worker profile / running mode / slide scope, starts child sessions, records the parent callback target, and reads handoffs after callback.
- Child session: reads shared context and handles the assigned package. In `production-light`, it directly writes briefs, prompts, generated images, `slides/slides.md` updates, and a compact handoff before callback.
+- Child session: reads shared context and handles the assigned package. In `production-light`, it directly writes briefs, prompts/specs, generated images or Remotion code/render outputs, `slides/slides.md` updates, and a compact handoff before callback.
 - Slide agent: optional worker for `test/audit`, explicit review, repair, exception handling, or intentionally isolated slide work. It handles exactly one slide when used.

 Operating lines:

 - `test/audit`: early project setup, core-shot proving, mechanism debugging, or failure investigation. Richer handoffs, ledgers, local review files, Agent workers, and parent inspection are allowed.
- `production-light`: normal production after core shots have reached the expected look and the user authorizes the switch in `project.md`. Keep the content line (`brief`, `prompt`, generated image, `slides/slides.md`) and compact recovery handoff; omit routine review, Agent workers, packet/return files, and audit records.
+- `production-light`: normal production after core shots or the core Remotion route have reached the expected look and the user authorizes the switch in `project.md`. Keep the content line (`brief`, `prompt/spec`, generated image or Remotion source/render, `slides/slides.md`) and compact recovery handoff; omit routine review, Agent workers, packet/return files, and audit records.

 Defaults:

 - Worker profile for current video / MV work: `video-slide-production`.
+- Worker profile for Remotion videos: `remotion-video-production`.
 - Design-gated slides: cover, PPT, science-video, title, explainer, and post-production-heavy pages require `sNN-vN-design.md` before `brief`.
 - Child session capacity: at most 8 ordinary production slides; reduce to 1-3 for identity-heavy, repair-heavy, core-risk, or context-expensive scopes.
 - Production-light topology: main session -> child session directly.
@ -274,8 +293,11 @@ The user controls task running mode:

 - `prompt only`: create briefs and prompts; do not generate images.
 - `prompt and generate`: create briefs and prompts, then generate images for the task's recorded slide scope.
+- `remotion plan only`: create Remotion execution specs and mappings; do not write source code.
+- `remotion implement`: write or edit Remotion source and run narrow source checks; do not render final MP4 unless the task also authorizes rendering.
+- `remotion render`: write or edit source as needed, run narrow source checks, and render the target MP4.

-Task-level generation instructions count as user approval for the named scope and mode. In production-light, generated images are recorded as production-pass outputs without per-image user review or routine final verification; the user can request an append-only retry later if a result is not usable. Core, repair, identity-heavy, or otherwise high-risk slides may still pause if the task plan says they require sampling or explicit review.
+Task-level generation or Remotion render instructions count as user approval for the named scope and mode. In production-light, generated images are recorded as production-pass outputs without per-image user review or routine final verification; the user can request an append-only retry later if a result is not usable. Remotion source/render work is code work and must run the task's narrow checks before closeout. Core, repair, identity-heavy, otherwise high-risk slides, or uncertain Remotion animation systems may still pause if the task plan says they require sampling or explicit review.

 For unattended production, the parent session repeats the callback loop from `docs/workflows/project-lifecycle.md`: update `production-run-status.md`, dispatch one child session, wait for callback, reload the minimal project/task state, then dispatch the next package if no stop condition is present. Exact parent context/token usage is not required for continuation because Codex may auto-compress the parent session; file-backed state is the authority.

@ -286,6 +308,18 @@ Visual-system rules:
 - Character continuity must use `visual-system/characters/character-reference-registry.md` and active character reference images.
 - Visible character identity cannot be generated from text-only prompts when identity matters.
 - Slide prompts must use the `gpt-image-2` Skill Advisor / host-native workflow.
+- Remotion implementation must use the installed Remotion Skill before code edits. Read `C:\Users\wangq\.agents\skills\remotion\SKILL.md`, then read relevant `rules/*.md` files for the task area.
+
+Remotion project rules:
+
+- Default source root: `projects/<project>/remotion/`.
+- Default deliverable: source code plus renderable MP4.
+- One-off project-specific Remotion source belongs in Video Workbench. Reusable products, reusable templates, deployable systems, and business software belong in `work-projects` unless the user explicitly keeps a prototype local.
+- The user has standing task-level permission for Remotion tasks to use `npx create-video`, npm dependency installation, and Remotion render commands when needed. Normal sandbox/network approval requests still apply at command time.
+- Keep source code, package files, and source assets in Git by default. Keep `node_modules`, caches, build outputs, and rendered media ignored by default.
+- Use `projects/<project>/remotion/public/` for assets referenced with Remotion `staticFile()` unless a task records a different local convention.
+- Use `projects/<project>/renders/` or `projects/<project>/exports/` for MP4 outputs unless the task records another render path.
+- For Remotion code, normal source edits are not append-only `sNN-vN-*` artifacts; use Git history and code review. Append-only rules still apply to Markdown control artifacts, scene specs, render checks, and generated media receipts.

 Review rules:

@ -293,6 +327,7 @@ Review rules:
 - Test/audit review defaults to local-only unless the task says otherwise.
 - Do not call Minimax MCP or other external vision/review services unless the user explicitly authorizes that task.
 - Generated production slides can be tracked in `tasks/<task-id>/task-log.md` and `slides/slides.md` instead of a full per-slide review.
+- Remotion render checks are source/render validation, not external visual review. Record the command, composition ID, output path, and result in the compact handoff or a `sNN-vN-render-check.md` file when the check is scene-specific.

 ## 9. Investigations

@ -317,11 +352,11 @@ Adjust only when the user specifies a different narration speed.

 Image and audio generation can be automated when available skills or APIs are appropriate.

-Video generation from images remains manual or semi-manual for now because output reliability varies.
+Video generation from images remains manual or semi-manual for now because output reliability varies. Remotion is a separate programmatic video path: use it when the accepted plan calls for React/TypeScript animation, scene composition, captions, charts, diagrams, or renderable MP4 output from source code.

 Editing and publishing are human-controlled unless explicitly automated later.

-Codex is the default prompt owner for final image prompts in this workspace, but final prompts must be produced through the `gpt-image-2` Skill workflow rather than direct freehand drafting. GPT V2 planning Markdown remains a source specification, not a final prompt.
+Codex is the default prompt owner for final image prompts in this workspace, but final prompts must be produced through the `gpt-image-2` Skill workflow rather than direct freehand drafting. GPT planning Markdown remains a source specification, not a final prompt.

 For project-bound image prompts, Codex must:

--- a/docs/context-handoff.md
+++ b/docs/context-handoff.md
@ -14,19 +14,21 @@ Skills Vault         = reusable automation skill supplier

 Video Workbench is a sibling of Writing Workbench. It should not be folded into article production because the target output, rhythm, assets, and evaluation criteria are different.

-Video Workbench keeps local project notes, execution outputs, and handoff records. It does not own canonical prompts, agent definitions, reusable automation source code, or production/business agent implementation for deployable systems.
+Video Workbench keeps local project notes, execution outputs, Remotion source for one-off project videos, and handoff records. It does not own canonical prompts, agent definitions, reusable automation source code, reusable Remotion products/templates, or production/business agent implementation for deployable systems.

 ## Collaboration Contract

 ```text
-GPT / ChatGPT   = stage 0-5 planning Markdown, visual-system design, page/shot skeletons
+GPT / ChatGPT   = stage 0-5 planning Markdown, visual-system design, page/shot skeletons, Remotion planning briefs
 User            = decisions and accepted input placement
-Codex           = stage 6+ execution planning, Prompt Advisor work, visual-system materialization, iteration, generated assets
+Codex           = stage 6+ execution planning, Prompt Advisor work, Remotion implementation, visual-system materialization, iteration, generated assets
 ```

-Most planning files should come from GPT V2 output accepted by the user and placed in `projects/<project>/intake/`. Codex-maintained files should be local execution aids and execution records.
+Most planning files should come from GPT output accepted by the user and placed in `projects/<project>/intake/`. Codex-maintained files should be local execution aids and execution records.

-Codex should not ask GPT V2 for final image prompts, Codex JSON, output paths, task lists, or image generation parameters. GPT V2 planning Markdown is a source specification, not an execution command.
+Codex should not ask GPT for final image prompts, Codex JSON, output paths, task lists, dependency commands, render commands, complete Remotion source code, or image generation parameters. GPT planning Markdown is a source specification, not an execution command.
+
+For Remotion videos, accepted planning from `强哥的策划导演3.0` can provide Remotion adaptation judgment, a Remotion style-system anchor, storyboard-to-component mapping, and a Remotion Implementation Brief. Codex turns that into local source code, tool commands, render checks, MP4 outputs, and compact project records.

 ## Project Shape

@ -51,6 +53,10 @@ projects/<project-id>/
    visual-system.md
    refs/
    iterations/
+  remotion/
+    package.json
+    src/
+    public/
  slides/
    slides.md
    s01/
@ -65,15 +71,17 @@ Use `docs/cases/` as the local Codex case-pattern library. It is allowed to extr

 `project.md` is a Codex-maintained small note. It should link to intake files and key local execution files instead of duplicating the outline, storyboard, page design, or execution plan.

-`intake/` holds GPT V2 stage 0-5 planning Markdown, user-provided source material, references, and other accepted input files.
+`intake/` holds GPT stage 0-5 planning Markdown, user-provided source material, references, and other accepted input files.

 `execution-plan.md` is the visual project's local execution control file. It records the current medium branch, visual-system materialization targets, small-batch iteration strategy, active `sNN` units, current round, and next decision. It does not repeat the project map from `project.md`.

 `visual-system/visual-system.md` records the current accepted visual system, materialization targets, iteration log, failure attribution, and next action. Character anchors, scene anchors, templates, diagrams, and backgrounds belong under `visual-system/`, not a separate `makeup-still/`.

+`remotion/` is the default source root for one-off project-specific Remotion videos. It should hold source code, package files, and source assets. Generated renders, caches, dependencies, and build outputs stay out of Git by default. Reusable Remotion products/templates belong in `work-projects` unless the user explicitly keeps a prototype local.
+
 `slides/slides.md` is the slide/page/shot execution fact table. It records each unit's source, batch, prompt version, image version, review state, acceptance state, and notes. It does not replace `execution-plan.md`.

-`slides/sNN/` keeps all files for one slide, page, or shot together: prompts, narration, generated images, generated audio, review notes, and local notes. Every file inside `sNN/` must carry the `sNN` prefix; iteration files use `sNN-vN-type.ext`, such as `s03-v1-prompt.md`.
+`slides/sNN/` keeps all files for one slide, page, shot, Remotion scene, or Remotion section together: prompts, narration, generated images, generated audio, component/spec notes, review notes, and local notes. Every file inside `sNN/` must carry the `sNN` prefix; iteration files use `sNN-vN-type.ext`, such as `s03-v1-prompt.md`.

 ## Investigations

@ -105,7 +113,9 @@ Produce a video usually under 10 minutes.
 Use about 220 Chinese characters per minute as the narration budget.
 ```

-Editing and publishing remain manual unless explicitly automated later. Image-to-video remains manual or semi-manual for now.
+Editing and publishing remain manual unless explicitly automated later. Image-to-video remains manual or semi-manual for now. Remotion videos are the exception for programmatic assembly: default project-specific source goes under `projects/<project>/remotion/`, and the default deliverable is source code plus a renderable MP4.
+
+The user has given standing task-level permission for Remotion tasks to use `npx create-video`, npm dependency installation, and Remotion render commands when needed. The agent must still use normal sandbox/network approval requests when a command requires escalation.

 ## Capability-Request Mode

--- a/docs/workflows/project-lifecycle.md
+++ b/docs/workflows/project-lifecycle.md
@ -30,8 +30,9 @@ The phase names stay stable; the concrete work differs by medium.
 | MV / shot-image package | `visual-system/`, style master, character registry, `slides/slides.md`. | Core shots, character consistency, reference-loading route, callback route. | Sequential slide image packages. |
 | Explainer video | Script structure, terminology, visual explanation rules, voice style, asset taxonomy. | One short script section, key diagram, sample narration audio, timing check. | Script sections, diagrams, narration audio, supporting slides. |
 | Slide-style presentation | Page system, component patterns, visual hierarchy, case/source map. | Representative pages and transition/page-density checks. | Page batches and export-ready assets. |
+| Remotion video | Remotion style-system anchor, storyboard-to-component mapping, `projects/<project>/remotion/` source root, composition plan, render target. | Scaffold or existing source check, representative scene/component, `npx remotion still` or equivalent narrow render check, MP4 render route. | React/TypeScript source updates, scene/component batches, renderable MP4 outputs, render receipts. |

-Do not copy GPT planning into duplicate project truth. The accepted `intake/` files remain sources; Codex-owned local execution rules live in `execution-plan.md`, `visual-system/`, `slides/`, `tasks/`, and medium-specific control files.
+Do not copy GPT planning into duplicate project truth. The accepted `intake/` files remain sources; Codex-owned local execution rules live in `execution-plan.md`, `visual-system/`, `slides/`, `tasks/`, `remotion/` when applicable, and medium-specific control files.

 ## Production Run State

@ -50,9 +51,11 @@ It records:
 - operating line;
 - running mode;
 - review mode;
+- worker profile;
 - package size or validation override;
 - total target scope;
 - completed units;
+- Remotion source root, composition ID, render target, and render output path when applicable;
 - active child session and callback target;
 - next unresolved cursor;
 - stop conditions;
@ -68,6 +71,7 @@ docs/workflows/slide-task-orchestration.md
 docs/workflows/slide-task-templates.md
 projects/<project>/project.md
 projects/<project>/execution-plan.md
+projects/<project>/visual-system/visual-system.md
 projects/<project>/slides/slides.md
 projects/<project>/tasks/<task-id>/production-run-status.md
 projects/<project>/tasks/<task-id>/task-log.md
--- a/docs/workflows/slide-task-orchestration.md
+++ b/docs/workflows/slide-task-orchestration.md
@ -3,10 +3,11 @@
 ## Status

 - Status: active workflow rule.
- Scope: local Video Workbench slide production tasks.
+- Scope: local Video Workbench slide, scene, page, and Remotion production tasks.
 - First profile: `video-slide-production`.
+- Remotion profile: `remotion-video-production`.

-In this workflow, `slide` is the execution unit. A slide may be a video frame unit, a PPT page, or another visual page-like unit. Current rules implement video/MV slide production only.
+In this workflow, `slide` is the historical execution unit. A unit may be a video frame unit, a PPT page, a Remotion scene, a Remotion section, or another visual page-like unit. Current rules implement image-led video/MV slide production and Remotion video production.

 ## Goals

@ -45,6 +46,36 @@ gpt-image-2 Advisor prompt/spec

 Garden/API use is expected to be `no` unless the user explicitly changes the project policy.

+Remotion profile:
+
+```text
+remotion-video-production
+```
+
+Use this profile when accepted planning material, the user, or the task says the video should be implemented in Remotion, when Codex writes or edits Remotion code, or when the expected deliverable is source code plus a renderable MP4.
+
+This profile uses:
+
+- accepted Remotion planning material in `intake/` as source specification, not as code;
+- `visual-system/` and the Remotion style-system anchor as production authority;
+- `slides/slides.md` as the scene/section fact table when the project decomposes into units;
+- `slides/sNN/` for scene specs, component mapping notes, and render-check receipts;
+- `projects/<project>/remotion/` as the default source root;
+- `C:\Users\wangq\.agents\skills\remotion\SKILL.md` and task-relevant Remotion Skill rule files before code edits;
+- narrow code/render validation, such as typecheck/build when available and `npx remotion still` or the task's render command.
+
+For Remotion work, `gpt-image-2` is used only when the Remotion project separately needs generated still assets. Remotion implementation itself is React/TypeScript source work, not image prompt work.
+
+Default Remotion source and output paths:
+
+```text
+projects/<project>/remotion/        # source
+projects/<project>/renders/         # rendered MP4s or still checks unless overridden
+projects/<project>/exports/         # final export package when needed
+```
+
+One-off project-specific Remotion source belongs in Video Workbench. Reusable Remotion products, reusable templates, deployable systems, and business software belong in `work-projects` unless the user explicitly keeps a prototype local.
+
 Reserved future profile:

 ```text
@ -53,12 +84,18 @@ ppt-page-production

 Future PPT tasks may use `docs/cases/` as an additional case-pattern knowledge base and may need a different page brief, prompt, and review structure. Do not apply `video-slide-production` rules blindly to PPT work; design that profile when the first real PPT task needs it.

-Every task plan must record:
+Every task plan must record the active profile, for example:

 ```text
 Worker Profile: video-slide-production
 ```

+or:
+
+```text
+Worker Profile: remotion-video-production
+```
+
 ## Operating Lines

 Use two lines:
@ -70,9 +107,9 @@ production-light

 `test/audit` is for early project setup, first visual-system proving, core-shot trials, workflow debugging, failure investigation, or explicit user-requested review. It may use richer handoffs, local review files, Agent workers, packets, returns, ledgers, parent-side inspection, completion gates, and longer narrative logs.

-`production-light` is for normal production after core shots have reached the expected look. It is the default production line. It keeps only:
+`production-light` is for normal production after core shots or the core Remotion route have reached the expected look. It is the default production line. It keeps only:

- content artifacts: `sNN-vN-brief.md`, `sNN-vN-prompt.md`, generated image when allowed, and `slides/slides.md`;
+- content artifacts: `sNN-vN-brief.md`, `sNN-vN-prompt.md` or `sNN-vN-remotion-spec.md`, generated image or Remotion source/render output when allowed, and `slides/slides.md`;
 - compact recovery records: child-session handoff and one-line task log entries.

 In production-light, review mode defaults to:
@ -109,7 +146,15 @@ projects/<project-id>/
      s05-v1-brief.md
      s05-v1-prompt.md
      s05-v1-image.png
+      s05-v1-remotion-spec.md
+      s05-v1-render-check.md
      s05-v1-review.md   # test/audit or explicit review only
+  remotion/
+    package.json
+    src/
+    public/
+  renders/
+    # generated render outputs ignored by default
 ```

 Rules:
@ -117,6 +162,8 @@ Rules:
 - `tasks/<task-id>/` records orchestration and agent communication.
 - `slides/sNN/` remains the durable home for slide artifacts.
 - In `slides/sNN/`, Markdown control/script artifacts are Git-tracked by default; generated image, audio, and video files are asset-layer outputs ignored by default and referenced from control files by path, status, and hash when needed.
+- `remotion/` is the default source root for project-specific Remotion videos. Source code, package files, and source assets are Git-tracked by default.
+- Remotion dependencies, caches, build outputs, and rendered media are ignored by default. Record render paths and status in control files instead of committing generated media unless the user promotes a named artifact.
 - `intake/` and `visual-system/` are source-authority areas and should stay in Git, including approved reference images.
 - `archive/` is for discarded images and discarded documents by default. Do not add archive contents to Git unless the user explicitly promotes a specific item back into a source-authority area.
 - Task records link to slide artifacts by path and should not duplicate full prompts.
@ -138,6 +185,8 @@ Default policy:

 This rule applies even when the previous run was malformed. Malformed output is still evidence about the workflow and should be superseded by a new version, not silently replaced.

+Remotion source files under `projects/<project>/remotion/` are ordinary code, not append-only generated artifacts. Edit them in place under Git while respecting the user's uncommitted changes. Do not delete or replace an existing Remotion scaffold just to get a clean project; inspect it and work with it. Append-only versioning still applies to scene specs, design docs, render-check receipts, generated media receipts, packets, returns, and handoffs.
+
 ## Session Roles

 ### Main Session
@ -199,6 +248,7 @@ Responsibilities:
 - read active `visual-system/` files needed for assigned slides;
 - resolve the next unused slide version and write one new `sNN-vN-brief.md` per assigned slide;
 - in `production-light`, write `sNN-vN-prompt.md`, generate `sNN-vN-image.png`, and update `slides/slides.md` directly;
+- for `remotion-video-production`, read the installed Remotion Skill and relevant rule files before code edits, write scene specs or component mapping notes as needed, update `projects/<project>/remotion/`, run narrow source/render checks, record composition IDs and render paths, and update `slides/slides.md` or task status directly;
 - in `production-light`, do not create slide-agent packets, do not start Agent/SubAgent workers, do not write return receipt files, and do not run a routine final verification pass;
 - in `production-light`, finish all project writes and narrow required checks before callback; do not continue writing after callback succeeds;
 - in `test/audit`, repair, or exception scopes, create one slide-agent packet per slide and start exactly one slide agent per slide, sequentially;
@ -207,6 +257,8 @@ Responsibilities:
 - write `child-session-handoff.md` with a compact parent-facing summary;
 - send the parent callback only after `child-session-handoff.md` exists.

+For Remotion work, the narrow source/render checks recorded in the task plan are part of producing code and MP4 output. They are required validation, not a routine visual-review pass.
+
 If a test/audit run needs an Agent and the runtime does not expose an Agent tool inside the child session, record this as a runtime fallback requirement in `child-session-handoff.md`. The main session may then start the slide agent as a fallback, but that run must be reported as a fallback orchestration test, not as a full child-session-owned slide-agent test.

 ### Slide Agent
@ -223,9 +275,11 @@ Responsibilities:

 - read the packet and brief;
 - follow the `gpt-image-2` Skill Advisor / host-native prompt workflow;
+- for Remotion packets, follow the installed Remotion Skill and task-relevant rule files instead of the `gpt-image-2` workflow;
 - preserve visual-system reference choices in `sNN-vN-prompt.md`;
 - write the new-version `sNN-vN-prompt.md`;
 - generate an image only when the packet allows generation;
+- write or edit Remotion code only when the packet explicitly uses `remotion-video-production`;
 - write `sNN-vN-review.md` only when required;
 - write a compact return receipt.

@ -247,6 +301,7 @@ The slide agent must not:
 - replace style-master guidance with ad hoc style text;
 - replace active character image references with text-only identity description when character continuity matters;
 - use archived style or character references unless the packet explicitly reopens them;
+- invent Remotion component style outside the accepted Remotion style-system anchor unless the task explicitly asks for a new pilot;
 - call Minimax MCP or external review tools by default.

 ## Running Modes
@ -284,6 +339,51 @@ child session writes brief

 Generation can happen later through a separate task.

+### Remotion Plan Only
+
+Use when the user wants Codex to translate accepted Remotion planning into local execution specs without writing source code.
+
+Flow:
+
+```text
+child session reads intake + visual-system Remotion anchor
+-> child session writes or updates scene specs / component mapping notes
+-> child session updates slides/slides.md or task status
+-> child session writes compact handoff and callbacks parent
+```
+
+No source code, npm command, or render command is allowed in this mode.
+
+### Remotion Implement
+
+Use when the user authorizes Remotion source work but does not request a final MP4 render.
+
+Flow:
+
+```text
+child session reads Remotion Skill and relevant rules
+-> child session creates or updates projects/<project>/remotion/
+-> child session writes or updates scene specs when needed
+-> child session runs narrow source checks recorded in the task plan
+-> child session records source paths, composition IDs, check result, and handoff
+```
+
+### Remotion Render
+
+Use when the user authorizes source work plus a renderable MP4.
+
+Flow:
+
+```text
+child session reads Remotion Skill and relevant rules
+-> child session creates or updates projects/<project>/remotion/
+-> child session runs narrow source checks
+-> child session runs the target Remotion still/render command
+-> child session records composition ID, output MP4 path, check result, and handoff
+```
+
+The user has given standing task-level permission for Remotion tasks to use `npx create-video`, npm dependency installation, and Remotion render commands when needed. The agent must still request sandbox or network approval at command time when required.
+
 ## Authorization Rule

 The user gives the main session concrete instructions for:
@ -291,6 +391,7 @@ The user gives the main session concrete instructions for:
 - running mode;
 - slide scope;
 - generation permission;
+- Remotion implementation or render permission;
 - operating line;
 - review or sampling expectations.

@ -298,6 +399,8 @@ If the task says `prompt only`, no image generation is allowed.

 If the task says `prompt and generate`, the task plan must record the authorized slide scope and exceptions. That task-level instruction is the generation approval for the named scope. In production-light, generated images are accepted into the production pass without per-image user review. Core, repair, identity-heavy, or otherwise high-risk slides may still require prompt-only handling or an explicit pause when the task plan says so.

+If the task says `remotion plan only`, no source edits or render commands are allowed. If the task says `remotion implement`, source edits and narrow source checks are allowed but final MP4 rendering is not required. If the task says `remotion render`, source edits, checks, and the target MP4 render are authorized for the named scope.
+
 ## Visual-System Requirements

 `visual-system/` is the production authority.
@ -309,6 +412,14 @@ For every slide:
 - record style module and active style references in the brief and prompt;
 - choose character references from `visual-system/characters/character-reference-registry.md` when a visible character matters.

+For every Remotion scene or section:
+
+- use the accepted Remotion style-system anchor from `visual-system/remotion/` or `visual-system/visual-system.md`;
+- preserve component style, information density, animation language, transition motifs, chart/flow/formula handling, and subtitle strategy from the accepted planning;
+- map GPT-provided component suggestions to real React/TypeScript components during Codex implementation;
+- read `C:\Users\wangq\.agents\skills\remotion\SKILL.md` and the relevant Remotion Skill rule files before code edits;
+- do not use CSS transitions or CSS animations for timeline-critical motion; use Remotion frame-based animation patterns from the Skill.
+
 For visible character slides:

 - text-only identity prompting is not sufficient;
@ -332,6 +443,8 @@ sNN-vN-design-en.md   # optional English companion; default design remains Chine
 sNN-vN-brief.md
 sNN-vN-prompt.md
 sNN-vN-image.png      # only when generation is allowed and succeeds
+sNN-vN-remotion-spec.md      # for Remotion scene/component intent when applicable
+sNN-vN-render-check.md       # for scene-specific Remotion still/render evidence when applicable
 sNN-vN-review.md      # only for test/audit or explicit review
 ```

@ -343,6 +456,10 @@ sNN-vN-review.md      # only for test/audit or explicit review

 `prompt` is an attachment to the brief. It should be runnable or easy to copy into another platform for cross-validation.

+`remotion-spec` is an execution-facing scene/component mapping for Remotion. It records source basis, composition/component target, text and data to render, animation intent, timing, assets, acceptance criteria, and which Remotion Skill rule files were read. It is not a replacement for React/TypeScript source.
+
+`render-check` records the command, composition ID, frame or render target, output path, and result for Remotion validation.
+
 `review` is absent by default in production-light. Create it only for test/audit, core trials, identity-heavy sampling, repair targets, explicit user requests, uncertain failures, or durable reference acceptance.

 When review is required or suppressed, the task plan or slide-agent packet should say so explicitly. Production-light tasks default to `review: none`; test/audit tasks default to local review unless the task plan disables it.
@ -358,10 +475,11 @@ The child session must:
 - assigned slide versions were resolved by scanning existing `slides/sNN/sNN-v*-*` files;
 - no target file was overwritten unless an explicit overwrite repair authorized that exact path;
 - each generated slide has a project image path recorded in `slides/slides.md` and the compact handoff;
+- each Remotion task records source root, composition ID, check command, check result, and render output path when rendering was authorized;
 - review files are absent when review mode is `none`;
 - the child session handoff exists before callback.

-Do not run a routine final verification pass in production-light. Do not compute hashes, build file inventories, inspect generated images, or write return receipts unless the task explicitly asks for that evidence or a problem occurs.
+Do not run a routine final verification pass in production-light. Do not compute hashes, build file inventories, inspect generated images, or write return receipts unless the task explicitly asks for that evidence or a problem occurs. For Remotion, the task's narrow source/render checks are required and should stay limited to the declared composition or render target.

 If any production-light closeout item fails, return `blocked`, `failed`, or `warning` and state what the main session should inspect. The main session then leaves lightweight production mode for that item and performs targeted inspection only.

@ -371,11 +489,13 @@ Use the fuller completion gate only in `test/audit`, repair, explicit review, or

 - append-only version resolution and no overwrite;
 - expected brief, packet, prompt, image when generation is allowed, and return files;
+- expected Remotion source files, composition registration, render-check receipt, and MP4 output when Remotion rendering is allowed;
 - review files are present when review was required and absent when review was disabled;
 - exactly one owning slide agent for that slide/version when Agents are used;
 - each slide agent return is recorded and the agent lifecycle is terminal / closed;
 - required visible-context image references were loaded or the run blocked before generation;
 - generated project images have recorded project paths and source paths; hashes are optional unless requested;
+- Remotion MP4 outputs have recorded project paths; hashes are optional unless requested;
 - the child session handoff contains a compact parent summary.

 ## Review Rule
--- a/docs/workflows/slide-task-templates.md
+++ b/docs/workflows/slide-task-templates.md
@ -26,6 +26,8 @@ Production-light form:
 - Running mode:
 - Review mode: none
 - Slides:
+- Remotion source root: `projects/<project>/remotion/` if applicable
+- Remotion render target: <composition / output path / not applicable>
 - Child session:
 - Callback target:
 - Write policy: append-only; overwrite allowed: no
@ -58,6 +60,8 @@ Use a longer plan only in `test/audit`.

 ## Visual-System Usage

+## Remotion Usage
+
 ## Generation Evidence Policy

 ## Rules
@ -100,6 +104,7 @@ Use this for any production request that may continue across multiple child sess
 - Project:
 - Lifecycle phase:
 - Line:
+- Worker profile:
 - Running mode:
 - Review mode:
 - Package size:
@ -112,6 +117,9 @@ Use this for any production request that may continue across multiple child sess
 - Completed units:
 - Active child session:
 - Next cursor:
+- Remotion source root:
+- Remotion composition ID:
+- Remotion render output:

 ## Callback Target

@ -125,6 +133,7 @@ Use this for any production request that may continue across multiple child sess
 - Resolved versions:
 - Handoff path:
 - Image paths:
+- Remotion output:
 - Problems:
 - Needs user:

@ -137,6 +146,7 @@ Field rules:
 - `Completed units` should list only units completed in the current production run, not every historical accepted unit.
 - `Next cursor` is the next unresolved unit in the authorized target set.
 - `User confirmation` says whether a clean callback may continue automatically or must pause for user visual confirmation.
+- Remotion fields stay blank or `not applicable` for image-only tasks. For `remotion-video-production`, record source root, active composition ID, and render output path when rendering is authorized.
 - `Latest Callback` is updated by the parent after callback. A child session must not update project files after sending the callback.
 - Do not paste prompt text, full file inventories, or image review notes into this file.

@ -155,6 +165,8 @@ Production-light form:
 - Assigned slides:
 - Running mode:
 - Review mode: none
+- Remotion source root:
+- Remotion checks:
 - Write policy: append-only; overwrite allowed: no
 - Shared context:
 - Visual-system files:
@ -176,6 +188,10 @@ Production-light form:
 - Review mode: none
 - Resolved versions:
 - Image paths:
+- Remotion source root:
+- Remotion composition IDs:
+- Remotion render outputs:
+- Check commands/results:
 - Problems:
 - Needs user:
 - Observed elapsed/context:
@ -187,7 +203,7 @@ Handoff field rules:
 - `Status` states `complete`, `blocked`, `failed`, or `warning`.
 - `Topology` is `direct-child` for normal production-light. Use `test/audit-with-agent` only when the task explicitly uses internal slide workers.
 - `Review mode` states `none` for production-light unless the task explicitly changed it.
- `Resolved versions` and `Image paths` are the main continuation surface.
+- `Resolved versions`, `Image paths`, and Remotion source/render/check fields are the main continuation surface.
 - `Observed elapsed/context` is optional and should be filled only when visible to the child or reported by the user.
 - `Parent summary` should be short enough that the parent can continue without opening prompt, ledger, or image files when status is clean.
 - Do not include full file inventories, narrative timelines, or copied prompt text in a production-light handoff.
@ -196,7 +212,7 @@ Callback payload rules:

 - Send the callback only after all allowed output files, `slides/slides.md` updates, narrow required checks, and `child-session-handoff.md` are complete.
 - The callback is the child session's final action. Do not update the handoff, run closeout checks, or write any project file after a successful callback.
- The production-light callback should include status, topology, resolved versions, handoff path, image paths if complete, and any problems or user decisions needed.
+- The production-light callback should include status, topology, resolved versions, handoff path, image paths or Remotion output paths if complete, and any problems or user decisions needed.
 - Do not add a routine final verification pass, hashes, or per-file inventory to a clean production-light callback.
 - If callback cannot be sent, record the failure in `Callback Status`; the parent may then use the task's recorded polling fallback or user intervention path.

@ -225,6 +241,8 @@ Use this only in `test/audit`, explicit review, repair, exception handling, or a

 ## Gpt-Image-2 Advisor Rule

+## Remotion Skill Rule
+
 ## Host Generation Evidence Rule

 ## Generation Rule
@ -238,6 +256,12 @@ Use this only in `test/audit`, explicit review, repair, exception handling, or a
 ## Return File
 ```

+Packet field rules:
+
+- Use `Gpt-Image-2 Advisor Rule` only for image-generation packets.
+- Use `Remotion Skill Rule` for `remotion-video-production` packets. It must name `C:\Users\wangq\.agents\skills\remotion\SKILL.md`, the task-relevant Remotion `rules/*.md` files to read, the source root, allowed source edits, and required check/render commands.
+- A packet must not authorize both image generation and Remotion rendering unless the task explicitly needs generated still assets inside a Remotion project.
+
 ## Slide Agent Return

 Use this only when a slide agent was actually launched. Production-light direct-child packages do not write slide-agent returns.
@ -350,6 +374,8 @@ Brief field rules:

 For Video Workbench image generation, the brief should treat `gpt-image-2` as prompt/spec Advisor only. It should not require or expect Garden/API execution unless the task plan explicitly changes the project policy.

+For Remotion production, the brief should point to the accepted Remotion planning source and Remotion style-system anchor. It should not include React/TypeScript source code; code belongs under `projects/<project>/remotion/`.
+
 ## Slide Prompt

 ```text
@ -382,6 +408,79 @@ Prompt field rules:
 - Do not put long project explanation, detailed source audit, review conclusions, or task orchestration history in normal prompt files.
 - For Video Workbench, final prompts may name local paths for traceability, but the prompt must also say that the actual reference image must be loaded into visible context before host image generation when identity/reference control matters.

+## Remotion Scene Spec
+
+Use this for `remotion-video-production` when a scene, section, or component needs an execution-facing mapping before or alongside source edits.
+
+```text
+# S-NN Remotion Spec V1
+
+## Status
+
+## Based On Sources
+
+## Scene / Section Function
+
+## Remotion Style Anchor
+
+## Composition / Component Target
+
+## Text / Data / Assets
+
+## Timing And Voiceover Relationship
+
+## Animation Intent
+
+## Component Mapping
+
+## Required Remotion Skill Rules
+
+## Acceptance Checklist
+
+## Source Handoff
+```
+
+Remotion spec field rules:
+
+- `Based On Sources` links to accepted `intake/` files, Remotion Implementation Briefs, and relevant visual-system files.
+- `Composition / Component Target` names the intended composition ID or component path when known. If unknown, say `to be decided by Codex implementation`.
+- `Text / Data / Assets` lists the text, values, diagrams, audio, images, or public assets the scene must render.
+- `Timing And Voiceover Relationship` records duration, frame range, narration beat, subtitle/caption relationship, or `not applicable`.
+- `Animation Intent` describes motion as explanation, not decoration.
+- `Required Remotion Skill Rules` lists `C:\Users\wangq\.agents\skills\remotion\SKILL.md` and any read `rules/*.md` files, such as `rules/video-layout.md`, `rules/timing.md`, `rules/transitions.md`, `rules/subtitles.md`, or `rules/audio.md`.
+- `Source Handoff` states what must be implemented under `projects/<project>/remotion/`.
+
+## Remotion Render Check
+
+Use this when a Remotion still, source check, or MP4 render was run.
+
+```text
+# S-NN Remotion Render Check V1
+
+## Status
+
+## Source Root
+
+## Composition ID
+
+## Commands
+
+## Output Paths
+
+## Result
+
+## Issues
+
+## Next Action
+```
+
+Render check field rules:
+
+- `Commands` records exact commands run, such as typecheck/build, `npx remotion still`, or `npx remotion render`.
+- `Output Paths` records still and MP4 output paths. Generated media stays ignored by default unless the user promotes a named file.
+- `Result` states pass, blocked, failed, or warning.
+- Keep this compact. Do not paste full command logs unless the error text is needed for diagnosis.
+
 Artifact version / write policy rules:

 - Use a new slide version for every new task execution. Do not reuse an existing `vN` just because an earlier version is incomplete.
@ -389,6 +488,7 @@ Artifact version / write policy rules:
 - The child session, and any slide-agent packet when one is used, must state `Overwrite allowed: no` by default.
 - Before writing any output path, check whether the target file exists. If it exists and overwrite is not explicitly authorized for that exact file, stop and return an artifact-collision blocker.
 - Do not backfill missing brief or prompt files for an old generated image as if they were original generation inputs. If backfill is needed for analysis, label it as reconstruction / audit, not as the original prompt.
+- Remotion source files under `projects/<project>/remotion/` are normal code and may be edited in place under Git. Do not replace an existing Remotion scaffold without inspecting it. Append-only versioning still applies to Markdown specs, render-check receipts, generated media receipts, packets, returns, and handoffs.

 ## Slide Review

--- a/projects/2026-06-29-搭上AI这条船/execution-plan.md
+++ b/projects/2026-06-29-搭上AI这条船/execution-plan.md
@ -0,0 +1,81 @@
+# Execution Plan
+
+## Current Medium Branch
+
+- Branch: internal PPT full-page visual image production.
+- Format: 16:9 horizontal.
+- Page count: 24 main pages, with possible backup pages later only if requested.
+- Production mode: full-page images first; final PPT assembly after page images are ready.
+- Prompt ownership: Codex owns final image prompts; GPT intake is source specification, not final prompt.
+- Image prompt workflow: `gpt-image-2` Advisor / host-native path only unless the user explicitly changes policy.
+
+## Current Execution Focus
+
+Active focus:
+
+1. P01 / `s01` cover and public-style sample passed user confirmation.
+2. P02 / `s02` first content-page sample passed user confirmation.
+3. Remaining page image production `s03-s24` completed through production-light child sessions.
+4. Next focus is optional PPT assembly, user visual sampling, or append-only page repair.
+
+## Operating Defaults
+
+| Item | Default |
+| --- | --- |
+| Line | production-light for remaining pages. |
+| Worker profile | pending PPT-page-production design; do not blindly reuse MV shot profile. |
+| Running mode now | prompt and generate, full-page image-led. |
+| Later likely running mode | same. |
+| Review mode later | none, unless user asks for sample review / repair. |
+| Write policy | append-only; overwrite allowed: no. |
+| Source authority | P1-P24 single-page design skeleton. |
+
+## Visual-System Alignment Scope
+
+The visual-system stage must settle:
+
+1. Global palette and atmosphere.
+2. Page family taxonomy.
+3. Motif rules: ship ticket, route, radar, compass, deck, port, double track, funnel / gate, ladder, training card.
+4. Text-in-image tolerance and simplification rules.
+5. Which pages should act as visual anchors for the whole deck.
+
+## Sample-Gate Plan
+
+The user approved this process:
+
+| Page | Why |
+| --- | --- |
+| P01 | Cover and public-style sample: tests the unified dark AI route-map base, ship ticket, route glow, title treatment, and meeting atmosphere. |
+| P02 | First content-page sample: tests whether the same base can carry ordinary page structure and meeting-positioning content. |
+
+Decision rule:
+
+- If P01 exposes a common base-style issue, revise `visual-system/`.
+- If P01 exposes a cover-only issue, revise P01 only.
+- If P02 exposes a content-page readability / structure issue that affects later pages, revise `visual-system/` and the style master.
+- If P02 exposes only a page-specific issue, revise P02 only.
+- After P01 and P02 pass, enter production for the remaining pages.
+
+## Current Round
+
+- R0: project files initialized.
+- R1: visual-system base accepted.
+- R2: P01 cover / public-style sample generated and user-confirmed pass.
+- R3: P02 content-page sample generated and user-confirmed pass.
+- R4: remaining pages formal production completed for `s03-s24`.
+
+## Open Decisions
+
+| Decision | Current Default |
+| --- | --- |
+| Final deck style | Full-page visual-image-led PPT. |
+| Base visual style | Unified deep blue-black AI route operation-map base across all pages; page families vary motifs lightly only. |
+| Text treatment | Text may be embedded in images; minor text defects acceptable. |
+| People / portraits | Avoid concrete people unless later explicitly needed. Use abstract business-route metaphors. |
+| Source charter | Fact-check backup only; not a main compression source. |
+| Public visual duplicates | Merge stage 3 / 4 / 5 visual guidance; treat repeats as reinforcement, not conflict. |
+
+## Next Decision
+
+If requested, assemble the completed full-page images into a PPT or run targeted append-only repair for any page the user marks unusable.
--- a/projects/2026-06-29-搭上AI这条船/handoff/2026-06-29-ai-ship-formal-production-handoff.md
+++ b/projects/2026-06-29-搭上AI这条船/handoff/2026-06-29-ai-ship-formal-production-handoff.md
@ -0,0 +1,205 @@
+# 2026-06-29 搭上 AI 这条船 Formal Production Handoff
+
+## Status
+
+- Project: `2026-06-29-搭上AI这条船`
+- Handoff purpose: next clean session starts formal production.
+- Current lifecycle phase: formal production setup.
+- Current workflow line: production-light for remaining pages.
+- Running mode: prompt and generate.
+- Review mode: none by default.
+- Write policy: append-only; overwrite allowed: no.
+
+User decision evidence:
+
+- user_decision: 2026-06-29 user confirmed `slides/s01/s01-v1-image.png` and `slides/s02/s02-v1-image.png` both passed.
+- user_decision: 2026-06-29 user requested this handoff and stated the next new session will start formal production.
+
+## Next-Session Load Order
+
+Load these files in order:
+
+1. `AGENTS.md`
+2. `projects/2026-06-29-搭上AI这条船/handoff/2026-06-29-ai-ship-formal-production-handoff.md`
+3. `projects/2026-06-29-搭上AI这条船/project.md`
+4. `projects/2026-06-29-搭上AI这条船/execution-plan.md`
+5. `projects/2026-06-29-搭上AI这条船/slides/slides.md`
+6. `projects/2026-06-29-搭上AI这条船/visual-system/visual-system.md`
+7. `projects/2026-06-29-搭上AI这条船/visual-system/style/style-prompt-master.md`
+8. `projects/2026-06-29-搭上AI这条船/visual-system/motifs/motif-registry.md`
+9. `projects/2026-06-29-搭上AI这条船/intake/阶段5：P1-P24 单页设计骨架.md`
+
+Read narrower intake files only if needed:
+
+- `intake/阶段5：页面清单骨架.md`
+- `intake/阶段5：全局页面规则.md`
+- `intake/阶段4：通用表达与视觉系统.md`
+- `intake/阶段3：搭上 AI 这条船PPT总纲.md`
+
+Backup fact-check source only:
+
+```text
+C:\Users\wangq\Documents\Codex\knowledge-vault\work\internal\OPC销售\AI智能化客户经营团队启动会议章程.md
+```
+
+Do not re-compress the charter into the main PPT. Use it only to verify a factual claim when needed.
+
+## Production Basis
+
+The user selected:
+
+> 整页视觉图为主。
+
+User-confirmed common visual base:
+
+> 每页底板风格完全统一；统一深蓝黑 AI 航线作战图底盘，按页面家族只做轻微母题变化。
+
+User-confirmed samples:
+
+| Unit | Page | Image | Status |
+| --- | --- | --- | --- |
+| `s01` | P01 / 搭上 AI 这条船 | `slides/s01/s01-v1-image.png` | user-confirmed pass |
+| `s02` | P02 / 今天这场会，不是再动员一次 | `slides/s02/s02-v1-image.png` | user-confirmed pass |
+
+When helpful, use these two images as visual calibration samples. If a future generation prompt needs actual image-reference behavior, load the local sample image into visible context before generation; a path written inside a prompt is traceability only.
+
+## Remaining Scope
+
+Remaining pages:
+
+```text
+s03-s24
+```
+
+Source authority:
+
+```text
+intake/阶段5：P1-P24 单页设计骨架.md
+```
+
+Do not reorder pages, rename page titles, or replace the core conclusion unless the user explicitly changes the deck.
+
+## Recommended First Production Package
+
+Default next package:
+
+```text
+s03-s08
+```
+
+Reason:
+
+- These are the next six pages in source order.
+- PPT pages are text/layout-heavy, so six pages is a conservative first formal package.
+- If the next session judges context and generation flow to be stable, later packages can broaden up to 8 pages.
+
+Suggested package sequence:
+
+1. `s03-s08`
+2. `s09-s16`
+3. `s17-s24`
+
+Adjust only if the user requests a different scope.
+
+## Production Rules For Next Session
+
+Use the lightweight route unless the user changes it:
+
+- main/new session may directly produce the package;
+- no routine review file;
+- no child session required for this project unless the next session chooses to use project-default batching;
+- no full completeness gate;
+- no external vision review;
+- no Minimax MCP or other external review services;
+- no source-charter expansion;
+- no prompt review pause unless the user asks.
+
+For each page in scope:
+
+1. Resolve next unused `sNN-vN` by scanning the page folder.
+2. Write `sNN-vN-brief.md`.
+3. Write `sNN-vN-prompt.md`.
+4. Generate `sNN-vN-image.png`.
+5. Update `slides/slides.md`.
+
+Review mode stays:
+
+```text
+none
+```
+
+Do not write `sNN-vN-review.md` unless the user explicitly asks for review, repair, or diagnosis.
+
+## Prompt / Generation Route
+
+Use `gpt-image-2` Skill Advisor / host-native flow.
+
+Mode result in this session:
+
+```json
+{
+  "mode": "B-or-C",
+  "recommendation": "host-or-advisor",
+  "garden_mode_enabled": false,
+  "has_api_key": false
+}
+```
+
+Expected route:
+
+```text
+brief -> prompt -> host image generation -> copy image into slides/sNN/ -> update slides.md
+```
+
+Do not use Garden/API paths unless the user explicitly changes policy.
+
+## Text Policy
+
+The user accepts minor text, font-size, and typo defects because this is an internal time-constrained meeting deck.
+
+Still follow these defaults:
+
+- one title;
+- one core judgment;
+- short labels;
+- reduce long lists before image generation;
+- preserve page title and core conclusion from P1-P24 skeleton.
+
+If a page's generated Chinese text is imperfect but the image direction is good, treat that as a possible PPT-layer repair later, not an automatic image failure.
+
+## File State At Handoff
+
+Created / active control files:
+
+- `project.md`
+- `execution-plan.md`
+- `visual-system/visual-system.md`
+- `visual-system/style/style-prompt-master.md`
+- `visual-system/motifs/motif-registry.md`
+- `slides/slides.md`
+- `tasks/2026-06-29-visual-system-alignment/task-log.md`
+
+Created sample artifacts:
+
+- `slides/s01/s01-v1-design.md`
+- `slides/s01/s01-v1-brief.md`
+- `slides/s01/s01-v1-prompt.md`
+- `slides/s01/s01-v1-image.png`
+- `slides/s02/s02-v1-design.md`
+- `slides/s02/s02-v1-brief.md`
+- `slides/s02/s02-v1-prompt.md`
+- `slides/s02/s02-v1-image.png`
+
+No review files were written for the sample run.
+
+## Stop / Escalation Conditions
+
+Stop and ask the user if:
+
+- generated pages drift away from the accepted P01/P02 style;
+- text becomes too broken for internal meeting use;
+- a page requires a public-style change rather than a page-local repair;
+- generation output suggests the deck is becoming customer proposal / government report / SaaS dashboard / cyberpunk tech poster;
+- there is an artifact collision because a target `sNN-vN-*` path already exists.
+
+Otherwise continue production under the approved lightweight rules.
--- a/projects/2026-06-29-搭上AI这条船/intake/阶段3：搭上
+++ b/projects/2026-06-29-搭上AI这条船/intake/阶段3：搭上
@ -0,0 +1,66 @@
+# 《搭上 AI 这条船：AI 智能化客户经营团队启动会》PPT 总纲 v1
+
+## 0. 总定位
+
+这套 PPT 不是销售动员，不是制度宣贯，也不是培训课件。
+
+它的定位是：
+
+> **把一份内部章程，转成一次“兄弟一起开新局”的共识会议。**
+
+会议要达成的不是“大家听我安排”，而是：
+
+> **我把完整方案摊出来，大家一起补充、校准、升级，然后立刻进入第一周训练。**
+
+原始章程的核心目的，是建立一支不是传统卖课、卖软件、卖项目的新型团队，而是围绕企业智能化，建立客户经营、认知升级、场景发现、商务推进和项目落地之间的协同体系。
+
+---
+
+## 1. PPT 核心主线
+
+这套 PPT 的主线建议定为：
+
+> **船票有了，航线有了，人也找对了。
+> 现在先上船，边学边打。**
+
+展开链路：
+
+```text
+为什么现在必须变
+↓
+为什么我不能再一个人顶全部
+↓
+为什么这次要找你们
+↓
+为什么 AI 时代给了外行上船机会
+↓
+我们卖的不是 AI 产品，而是一条企业智能化路径
+↓
+客户怎么判断，不能乱推系统
+↓
+团队怎么分工，你们怎么成长
+↓
+先记住三条边界
+↓
+进入一个月培训
+↓
+第一周先形成自己的 AI 训练目标
+```
+
+---
+
+## 2. 45 分钟会议节奏
+
+| 模块        |    时间 | 作用                             |
+| --------- | ----: | ------------------------------ |
+| 开场定调      |  3 分钟 | 说明今天不是动员，是共识会                  |
+| 为什么必须变    |  7 分钟 | 讲清行业焦虑、AI机会、我的角色变化             |
+| 为什么找你们    | 10 分钟 | 重点建立信心：Agentic 技术平权，老销售经验变成新优势 |
+| 我们到底卖什么   |  8 分钟 | 讲三级服务路径：培训、顾问、系统               |
+| 客户怎么筛     |  5 分钟 | 讲客户适配判断模型                      |
+| 团队怎么协作    |  6 分钟 | 讲团队结构、五个角色、魏征机制                |
+| 怎么成长、怎么训练 |  6 分钟 | 讲 L0-L4、一个月培训、第一周目标            |
+
+现场可以随时打断交流，所以 PPT 本身不要做成“密集讲稿”，而要做成**每页一个判断，每页一个停留点**。
+
+---
--- a/projects/2026-06-29-搭上AI这条船/intake/阶段4：通用表达与视觉系统.md
+++ b/projects/2026-06-29-搭上AI这条船/intake/阶段4：通用表达与视觉系统.md
@ -0,0 +1,92 @@
+# 阶段4：通用表达与视觉系统
+
+## 1. 表达任务
+
+这套 PPT 要让参会者形成四个感觉：
+
+1. **兴奋感**：AI 这条船不是口号，已经有船票了。
+2. **信心感**：他们不是因为懂技术才被需要，而是因为懂人、懂局、懂客户。
+3. **秩序感**：兄弟一起开新局，但未来必须公司化、流程化。
+4. **行动感**：会后不空谈，马上进入第一周训练。
+
+---
+
+## 2. 语言风格
+
+建议使用：
+
+> **自己人口吻 + 公司化骨架 + 战斗前夜的兴奋感。**
+
+语言特点：
+
+* 用“我”和“我们”，不要过度组织化。
+* 判断要直接，不绕。
+* 少讲抽象概念，多讲“为什么是你们”。
+* 规则轻讲，训练重讲。
+* 技术判断可以保留原文锐度。
+
+关键句式：
+
+> **先上船，再学怎么开船。**
+> **第一阶段不要求懂 AI，要求会带着 AI 工作。**
+> **老销售的能力没有过时，只是换了发动机。**
+> **技术平权，不是人人都当程序员，而是外行也能参与探索。**
+
+---
+
+## 3. 统摄性隐喻
+
+核心隐喻：
+
+> **搭上 AI 这条船。**
+
+隐喻展开：
+
+| 隐喻   | 对应内容                 |
+| ---- | -------------------- |
+| 船票   | AI 时代的新机会已经出现        |
+| 上船   | 有意愿的人先进入训练和协作        |
+| 航线   | 培训 → 顾问 → 系统         |
+| 船员分工 | 专家、客户经营、AI业务工程师、工程底座 |
+| 雷达   | 客户判断、风险识别、机会发现       |
+| 罗盘   | 方法论、客户适配模型、承诺边界      |
+| 训练营  | 一个月培训，第一周 AI 习惯训练    |
+
+---
+
+## 4. 视觉基调
+
+不做人像，不做复杂插画。
+
+建议风格：
+
+> **深蓝黑商务底 + 航线图 + 雷达光效 + 金色/青色重点词 + 大模型图。**
+
+视觉关键词：
+
+* 深海蓝
+* 航线
+* 雷达扫描
+* 罗盘
+* 船票
+* 甲板
+* 港口
+* 路线图
+* 作战地图
+* 分层阶梯
+
+页面文字可以直接进入图中，因为这是内部会议使用，主要是替代长文阅读。
+
+---
+
+## 5. 固定页面模板
+
+建议固定 5 类页面：
+
+1. **封面页**：大标题 + 航线背景。
+2. **判断页**：一句大判断 + 3 个支撑点。
+3. **模型页**：三级路径、客户适配、L0-L4 等。
+4. **对比页**：传统销售 vs AI客户经营，传统工程师 vs Agentic工程师。
+5. **行动页**：第一周训练目标、会后作业。
+
+---
--- a/projects/2026-06-29-搭上AI这条船/intake/阶段5：P1-P24
+++ b/projects/2026-06-29-搭上AI这条船/intake/阶段5：P1-P24
--- a/projects/2026-06-29-搭上AI这条船/intake/阶段5：全局页面规则.md
+++ b/projects/2026-06-29-搭上AI这条船/intake/阶段5：全局页面规则.md
@ -0,0 +1,60 @@
+# 一、全局页面规则
+
+## 1. 页面气质
+
+这套 PPT 的气质是：
+
+> **兄弟一起开新局，但要按公司化打法打仗。**
+
+不是客户汇报，不是正式课件，不是制度宣贯。
+
+它应该像一张**上船前的作战地图**。
+
+---
+
+## 2. 统一视觉系统
+
+建议统一采用：
+
+> **深蓝黑商务底 + AI 航线图 + 雷达光效 + 青色/金色高亮关键词 + 大模型图。**
+
+视觉元素：
+
+* 船票；
+* 航线；
+* 罗盘；
+* 雷达；
+* 甲板；
+* 港口；
+* 双轨；
+* 阶梯；
+* 作战地图；
+* 客户旅程。
+
+---
+
+## 3. 页面文字策略
+
+因为这是内部交流，且你明确说 PPT 本质上是替代看文本，所以：
+
+* 页面可以直接承载文字；
+* 不用过度追求留白；
+* 但每页必须有一个大判断；
+* 每页最多一个主模型；
+* 不要把原文整段复制进页面。
+
+执行时，每页建议采用：
+
+> **一句大标题 + 一句核心判断 + 一个模型 / 三个要点 / 一张关系图。**
+
+---
+
+## 4. 讲述备注策略
+
+你自己讲，不需要完整备注。
+
+每页只给：
+
+> **讲述关键词 / 过渡句 / 容易讲散的提醒。**
+
+---
--- a/projects/2026-06-29-搭上AI这条船/intake/阶段5：页面清单骨架.md
+++ b/projects/2026-06-29-搭上AI这条船/intake/阶段5：页面清单骨架.md
@ -0,0 +1,140 @@
+
+# 阶段5：页面清单骨架 v1
+
+## 主 PPT：24 页
+
+|  页码 | 页面标题                       | 页面功能    | 核心结论                                  | 页面结构 / 视觉方向                                                | 讲述关键词                    |
+| --: | -------------------------- | ------- | ------------------------------------- | ---------------------------------------------------------- | ------------------------ |
+| P01 | 搭上 AI 这条船                  | 封面定调    | 船票已经有了，今天我们正式上船                       | 深蓝航线背景 + 大标题 + 副标题“AI智能化客户经营团队启动会”                         | 不是动员，是开局                 |
+| P02 | 今天这场会，不是再动员一次              | 会议定位    | 大家已经有意愿，今天是把方案摊开形成共识                  | 左侧“不是”，右侧“是”                                               | 共识、补充、升级                 |
+| P03 | 我们要建的，不是传统销售团队             | 核心定义    | 我们要建的是 AI 智能化客户经营团队                   | 中心大词“客户经营团队”，四周围绕：客户、认知、场景、商务、落地                           | 不卖话术，不卖热闹                |
+| P04 | 为什么现在必须变                   | 现实压力    | 旧行业下滑，AI 新船靠岸，不能再观望                   | 一边是下滑曲线，一边是航船靠岸                                            | 焦虑不是坏事，是上船动力             |
+| P05 | 我这半年打通了什么                  | 强哥方法论背书 | 产品力、方法论、培训路径已经准备到可以带队开局               | 三段式：产品力 / 方法论 / 培训体系                                       | 不是临时起意                   |
+| P06 | 我的角色已经变了                   | 角色转型    | 我不能再长期同时承担专家、商务、销售、方案和客户维护            | “全能型老板”到“学者型实践专家”的箭头                                       | 过去靠我顶，未来靠体系打             |
+| P07 | 过去是我一个人往前推，未来要靠团队一起跑       | 组织升级    | 个人驱动要升级为团队、流程、工具、方法论驱动                | 单点发动机 → 多引擎船队                                              | 我不是退出，是换位置               |
+| P08 | 为什么不是传统销售                  | 销售身份重定义 | 这件事不是把 AI 卖出去，而是把客户带上正确路径             | 传统销售 vs 客户经营负责人对比                                          | 不是卖东西，是经营客户旅程            |
+| P09 | 为什么这次找你们                   | 信心建立    | AI 客户经营第一阶段，最重要的是懂人、懂局、懂客户            | 三个大词：识人 / 识局 / 识势                                          | 你们的老经验没有过时               |
+| P10 | Agentic 技术平权：外行也能上船        | 核心信心页   | AI 时代，低技术包袱的人反而更容易参与探索                | “技术门槛下降，业务理解上升”的交叉图                                        | 不懂代码，不等于不能做 AI           |
+| P11 | 传统软件工程师和 Agentic 工程师，是两个物种 | 反直觉判断   | 这不是技术栈升级，而是能力系统变化                     | 左右对比：确定性工程 vs 不确定性探索                                       | 这页是给大家信心，不是贬低技术          |
+| P12 | 第一阶段要求：不是懂 AI，而是会带着 AI 工作  | 学习门槛降维  | 先养成 AI 工作习惯，再谈客户经营能力升级                | “事前问 AI / 事后用 AI 复盘”循环图                                    | 先会用，再会卖，再会带客户            |
+| P13 | 我们卖的不是 AI 产品，是一条企业智能化路径    | 产品定义    | 客户真正需要的是从想用 AI 到用好 AI 的路径             | 航线图：想用 AI → 能用 AI → 用好 AI                                  | 产品不是单点，是路径               |
+| P14 | 三级服务路径：培训 → 顾问 → 系统        | 必讲模型 1  | 培训是入口，顾问是筛选，系统是重服务                    | 三段航线 / 三层阶梯                                                | 不是所有客户都直接上系统             |
+| P15 | 培训可以宽，顾问要筛，系统必须严           | 客户推进原则  | 宽入口、严系统，避免做错客户                        | 漏斗图：培训大口径 → 顾问筛选 → 系统少数                                    | 做错项目比少做项目更伤              |
+| P16 | 客户适配判断模型                   | 必讲模型 2  | 判断客户至少看四件事：基础、认知、场景、权力                | 四象限罗盘：基础建设 / 组织认知 / 业务场景 / 权力结构                            | 客户能不能做，比想不想做更重要          |
+| P17 | 双轨体系：先探索价值，再工程兑现           | 必讲模型 3  | 探索轨负责发现价值，工程轨负责兑现价值                   | 两条轨道图：探索轨在前，工程轨在后                                          | 眼下先拉上船，系统建设后面再细化         |
+| P18 | 新团队协作结构                    | 必讲模型 4  | 未来不是老板+销售+技术，而是专家、客户经营、AI业务工程师、工程底座协作 | 中心结构图：强哥 / 客户经营 / AI业务工程师 / 工程底座                           | 每个人不是什么都干，而是各就各位         |
+| P19 | 我们是帮客户上船的人                 | 精神锚点降维  | 客户不知道怎么上 AI 这条船，我们就是带路的人              | 五个角色围绕一艘船：入口、雷达、翻译器、推进器、防火墙                                | 找到客户、听懂客户、判断客户、推动客户、守住边界 |
+| P20 | 魏征机制：你们也是我的现实校准器           | 关系机制    | 你们不只是销售，也要提醒我、校准我、保护会议节奏              | “会前提醒 / 会中拉回 / 会后复盘”三段图                                    | 不是打断我，是保护节奏              |
+| P21 | L0-L4 成长路径                 | 必讲模型 5  | 头衔不是先给的，是通过训练和行为拿到的                   | 阶梯图：L0渠道伙伴 → L1 AI商务助理 → L2客户经营负责人 → L3商务负责人 → L4 AI前端部署顾问 | 先定位，再成长                  |
+| P22 | 今天先记住三条                    | 轻规则页    | 不装专家，不乱承诺，不把不成熟客户直接推系统                | 三张大卡片                                                      | 规则以后慢慢细化，今天先钉住底线         |
+| P23 | 一个月训练：先成为 AI 增强型客户经营人员     | 培训总览    | 一个月后，大家至少会用 AI 做准备、复盘和客户判断            | 四周训练路线：交朋友 / 懂客户 / 知边界 / 模拟拜访                              | 不是培养技术专家，是培养 AI 增强型商务    |
+| P24 | 第一周目标：写出自己的 AI 训练目标        | 行动收束    | 会后第一件事：每个人形成自己的第一周 AI 训练目标            | 作业页：我准备用 AI 解决什么、每天何时用、会前会后怎么用、一周后交什么                      | 先从自己开始，把 AI 用起来          |
+
+---
+
+## 3. 重点页面处理建议
+
+### P10-P12 是本次 PPT 的信心核心
+
+这里不要讲成技术课。
+
+核心不是解释 Agentic，而是让他们明白：
+
+> **信息化时代，他们只能在外围。
+> Agentic 时代，他们可以进入前场。**
+
+这一组页面要打出气势：
+
+> 不是你们不懂技术，所以不适合。
+> 恰恰是因为这件事一开始不是写代码，而是理解人、理解业务、理解场景，所以你们适合先上船。
+
+原文章程中也明确指出，销售第一阶段的要求不是懂 AI，而是会带着 AI 工作；同时，Agentic 工程面对的是目标不清晰、路径不确定、业务上下文复杂的问题，这正是传统确定性工程之外的新能力空间。
+
+---
+
+### P14-P16 是业务路径核心
+
+这三页负责把“我们到底卖什么”讲清楚。
+
+一句话：
+
+> **我们不是见客户就卖系统。
+> 我们先用培训打开入口，用顾问筛出价值，最后才谈系统。**
+
+原文章程中对客户分层已经有清晰判断，尤其是“培训可以宽，顾问要筛，系统必须严”这句话，应该保留为主视觉金句。
+
+---
+
+### P18-P21 是团队关系核心
+
+这四页要让他们明白：
+
+> **不是今天给大家封官，而是给大家一条成长路径。**
+
+重点讲两个模型：
+
+1. **团队协作结构**：强哥不再什么都亲自扛，客户经营团队要站到前台。
+2. **L0-L4 成长路径**：每个人从自己的现实条件出发，不必一步到位。
+
+原文章程里，启动会议要达成的共识包括：不是传统销售团队、销售要成为 AI 增强型商务人员、强哥从全能型老板转为学者型首席专家、客户经营负责人要做读人识局和“魏征”等。
+
+---
+
+## 4. 主金句库
+
+可分散放到页面标题、副标题或结束页：
+
+1. **船票现在有了，先上船。**
+2. **我们卖的不是 AI 产品，而是一条企业智能化路径。**
+3. **第一阶段不要求懂 AI，要求会带着 AI 工作。**
+4. **老销售的能力没有过时，只是换了发动机。**
+5. **技术平权，不是人人当程序员，而是外行也能参与探索。**
+6. **培训可以宽，顾问要筛，系统必须严。**
+7. **客户不知道怎么上 AI 这条船，我们就是带路的人。**
+8. **头衔不是先给的，是训练和行为拿到的。**
+9. **规则可以慢慢细化，但底线今天先钉住。**
+10. **先把 AI 用到自己身上，再带客户上 AI 这条船。**
+
+---
+
+## 5. 不建议进入主 PPT 的内容
+
+这些内容不是不重要，而是不适合这次启动会展开：
+
+* 佣金分成；
+* 报价规则；
+* 详细权限；
+* 四级承诺矩阵细则；
+* 禁止话术大全；
+* 红线考试；
+* 销售工具模板；
+* 项目交接模板；
+* 智能体清单明细；
+* 培训结业标准。
+
+它们适合后续变成：
+
+> **培训材料 / 章程附录 / 流程模板 / 工作手册。**
+
+---
+
+## 6. 备份页建议
+
+可放 4 页备份，不主动讲，现场有人问再打开：
+
+1. 章程原文目录；
+2. 一个月培训详细安排；
+3. 客户分层表；
+4. 四级承诺矩阵简版。
+
+---
+
+## 7. 导演审查
+
+这一版最重要的调整是：
+
+> **把“传统软件工程师和 Agentic 工程师是两个物种”前置为信心建立页，而不是后置为双轨分工页。**
+
+因为这次启动会的第一任务不是讲未来系统建设，而是让这些销售老兵意识到：
+
+> **AI 时代不是把他们挡在门外，而是给了他们重新进场的机会。**
--- a/projects/2026-06-29-搭上AI这条船/project.md
+++ b/projects/2026-06-29-搭上AI这条船/project.md
@ -0,0 +1,103 @@
+# 2026-06-29-搭上AI这条船
+
+## Status
+
+- Status: P01-P24 full-page image production completed via production-light child sessions.
+- Medium type: 16:9 internal kickoff meeting PPT.
+- Output direction: full-page visual-image-led deck.
+- Lifecycle phase: page image production complete.
+- Workflow phase: production-light run completed for remaining pages.
+- Review default for later production: none, unless the user requests sampling or repair.
+- Current execution focus: optional PPT assembly, user visual sampling, or append-only repairs if requested.
+
+## Meeting Positioning
+
+This deck is for an internal kickoff speech. It is not an archive version of the charter, not a customer proposal, and not a formal training courseware package.
+
+The deck should create shared confidence, make the team feel they can board the AI route now, and move directly into first-week AI training goals.
+
+Core sentence:
+
+> 船票有了，航线有了，人也找对了。现在先上船，边学边打。
+
+## Output Format Decision
+
+The user selected:
+
+> 整页视觉图为主。
+
+Execution meaning:
+
+- Each PPT page may be treated as a 16:9 full-page visual image.
+- Page text may appear directly inside the generated image.
+- Minor typography, font-size, and typo imperfections are acceptable because this is a time-constrained internal meeting deck.
+- The deck should prioritize atmosphere, momentum, page-level judgment, and visual memory over fine editable text polish.
+- If a page has too much text for reliable image rendering, the working default is to simplify visible text to title, core judgment, and short labels.
+
+## Source Hierarchy
+
+Main PPT authority:
+
+1. `intake/阶段5：P1-P24 单页设计骨架.md`
+2. `intake/阶段5：页面清单骨架.md`
+3. `intake/阶段5：全局页面规则.md`
+4. `intake/阶段4：通用表达与视觉系统.md`
+5. `intake/阶段3：搭上 AI 这条船PPT总纲.md`
+
+Backup fact-check source only:
+
+```text
+C:\Users\wangq\Documents\Codex\knowledge-vault\work\internal\OPC销售\AI智能化客户经营团队启动会议章程.md
+```
+
+The charter must not be re-compressed into the main PPT. Use it only if a factual claim in the intake files needs verification.
+
+## Accepted Intake
+
+| File | Role | Use |
+| --- | --- | --- |
+| `intake/阶段3：搭上 AI 这条船PPT总纲.md` | meeting story | Main emotional arc, timing, and core story chain. |
+| `intake/阶段4：通用表达与视觉系统.md` | expression / visual system | Tone, language, visual motifs, page family guidance. |
+| `intake/阶段5：P1-P24 单页设计骨架.md` | page-level authority | Canonical page order, page titles, conclusions, and per-page visual direction. |
+| `intake/阶段5：全局页面规则.md` | global page rules | Deck tone, text strategy, and shared visual elements. |
+| `intake/阶段5：页面清单骨架.md` | page table | Compact page list and key-page treatment notes. |
+
+## Current Visual Direction
+
+> 深蓝黑商务底 + AI 航线图 + 雷达光效 + 青色/金色高亮关键词 + 船票 / 罗盘 / 路线图 / 作战地图视觉母题。
+
+The feel should be:
+
+> 兄弟一起开新局，但要按公司化打法打仗。
+
+Accepted base direction:
+
+> 每页底板风格完全统一；统一深蓝黑 AI 航线作战图底盘，按页面家族只做轻微母题变化。
+
+## Key Local Execution Files
+
+| File | Purpose |
+| --- | --- |
+| `execution-plan.md` | Current execution control and next decision. |
+| `visual-system/visual-system.md` | Merged shared visual system. |
+| `visual-system/style/style-prompt-master.md` | Reusable style and prompt-language master for later page prompts. |
+| `visual-system/motifs/motif-registry.md` | Motif-to-page-family registry. |
+| `slides/slides.md` | P01-P24 page execution status table. |
+| `tasks/2026-06-29-visual-system-alignment/task-plan.md` | Current setup and visual-system alignment task. |
+
+## Production Handoff
+
+Current handoff for the next clean session:
+
+```text
+handoff/2026-06-29-ai-ship-formal-production-handoff.md
+```
+
+User decision evidence:
+
+- 2026-06-29: user confirmed both generated sample images passed: `slides/s01/s01-v1-image.png` and `slides/s02/s02-v1-image.png`.
+- 2026-06-29: formal production completed for `s03-s24` across three direct child sessions; review mode `none`.
+
+## Next Action
+
+Use the completed page images as the production-pass source for optional PPT assembly. If the user finds a page unusable, repair it append-only as a new `sNN-vN` version.
--- a/projects/2026-06-29-搭上AI这条船/slides/s01/s01-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s01/s01-v1-brief.md
@ -0,0 +1,126 @@
+# S-01 Slide Brief V1
+
+## Status
+
+- Status: ready for prompt.
+- Source page: P01.
+- Output type: 16:9 full-page visual image.
+- Running mode: prompt and generate.
+- Review mode: none.
+- Write policy: append-only; overwrite allowed: no.
+
+## Source Basis
+
+- `../../intake/阶段5：P1-P24 单页设计骨架.md`
+- `../../visual-system/visual-system.md`
+- `../../visual-system/style/style-prompt-master.md`
+- `../../visual-system/motifs/motif-registry.md`
+- `s01-v1-design.md`
+
+Backup fact-check source recorded but not used:
+
+```text
+C:\Users\wangq\Documents\Codex\knowledge-vault\work\internal\OPC销售\AI智能化客户经营团队启动会议章程.md
+```
+
+## Slide Function
+
+封面定调，同时作为整套 PPT 的公共视觉样片。它要让参会者第一眼明白：这不是随便聊聊，而是正式开局。
+
+## First Read
+
+第一眼应该读到：
+
+> 搭上 AI 这条船。
+
+第二眼读到：
+
+> 船票已经有了，今天我们正式上船。
+
+第三眼读到深蓝黑 AI 航线作战图、发光航线、抽象船票、远方港口灯光。
+
+## Risk Class
+
+- Risk class: core motif / title-safe cover.
+- Public-style risk: high, because this page tests the common deck base.
+- Page-specific risk: medium, because title and ship-ticket placement can be adjusted locally.
+
+## Reference Plan
+
+- Actual visual reference image: none.
+- Style source: `visual-system/style/style-prompt-master.md`
+- Motif source: `visual-system/motifs/motif-registry.md`
+- Active style module: Module A, Opening / Consensus.
+- Core motifs: AI 航线, 船票, 港口 / 灯塔, 雷达.
+- Generation path: host-native image generation; Garden/API expected: no.
+
+## Visual Requirements
+
+- 16:9 horizontal full-page cover image.
+- Deep navy-black AI maritime route-map base.
+- Cyan glowing route from lower-left / dark origin toward upper-right / far port.
+- Abstract AI boarding ticket as a foreground signal, gold edge and cyan data marks.
+- Far port / lighthouse warm-gold signal light.
+- Subtle radar scan, compass marks, coordinate grid, and navigation lines.
+- No real people, no real meeting room, no literal large ship.
+
+## Character Handling
+
+No visible character. Do not add faces, silhouettes of attendees, speakers, or crew.
+
+## Composition / Scale
+
+- Main title: central-left or mid-upper area, dominant.
+- Route line: lower-left to upper-right, behind title but readable.
+- Ticket: lower-right preferred, 12%-18% of frame, secondary to title.
+- Port light: upper-right / far-depth, small but legible.
+- Radar / compass: background layer only.
+
+## Light / Color / Texture
+
+- Base: deep blue-black, dark navy, cold slate gray.
+- Signal: cyan for route / AI / radar, restrained gold for ticket / opportunity / port.
+- Title: white or pale-blue, with subtle cyan/gold accent on `AI` if stable.
+- Texture: cinematic business, map glass, subtle scan lines, no cyberpunk excess.
+
+## Text Plan
+
+Use in-image text. Minor text defects are acceptable, but keep text short.
+
+Required text priority:
+
+1. `搭上 AI 这条船`
+2. `AI 智能化客户经营团队启动会`
+3. `先上船，再学怎么开船。`
+
+Optional small label:
+
+- `内部共识会议 / 启动培训前置会`
+
+If text rendering is unstable, keep the main title and subtitle; bottom and corner text can be repaired later in PPT.
+
+## Forbidden Drift
+
+Avoid:
+
+- government meeting cover;
+- customer proposal cover;
+- generic tech summit poster;
+- cyberpunk neon city;
+- cute cartoon ship / pirate ship / fantasy ship;
+- literal cruise ship hero;
+- humanoid robots or AI faces;
+- dense dashboard UI;
+- text too small or paragraph-like.
+
+## Acceptance Checklist
+
+- The cover feels like a formal internal kickoff, not a customer-facing deck.
+- The AI route / boarding ticket metaphor is clear.
+- The unified deep navy-black operation-map base is usable for later pages.
+- The title is the first visual read.
+- The image has excitement and order at the same time.
+
+## Prompt Handoff
+
+Create a host-native image prompt using the active Module A language and this brief. Do not use external reference images. Record `mode: B-or-C / host-or-advisor`. Generate directly in the main session after prompt creation, with no review file and no child session.
--- a/projects/2026-06-29-搭上AI这条船/slides/s01/s01-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s01/s01-v1-design.md
@ -0,0 +1,169 @@
+# S-01 Slide Design V1
+
+## Status
+
+- Status: draft for user review.
+- Source page: P01.
+- Page title: 搭上 AI 这条船.
+- Role: cover page and public-style sample.
+- Line: test/audit sample.
+- Write policy: append-only; overwrite allowed: no.
+- Next gate: user confirms design before brief / prompt / image generation.
+
+## Design Purpose
+
+本页既是封面，也是整套 PPT 的公共视觉样片。它要一次性验证三件事：
+
+1. 统一深蓝黑 AI 航线作战图底板是否成立；
+2. “AI 船票 / 上船 / 航线”这个核心隐喻是否一眼能看懂；
+3. 封面是否有内部启动会的兴奋感，而不是政府汇报、客户方案或培训课件。
+
+如果本页失败，需要先区分原因：公共底板问题回改 `visual-system/`；封面局部构图问题只改 P01。
+
+## Discussion Summary
+
+用户已确认整套 PPT 采用“整页视觉图为主”，每页底板风格完全统一。公共方向为：
+
+> 统一深蓝黑 AI 航线作战图底盘 + 按页面家族轻微变化母题。
+
+P01 作为第一张样片，优先测试公共底板、标题字、青色 / 金色高亮、航线、雷达、船票与港口远光的组合。
+
+## Visual Concept
+
+画面是一张“上船前的 AI 作战航线图”。
+
+底层是深蓝黑的数字海面 / 航海地图，带有非常克制的网格、坐标线、罗盘刻度和雷达扫描光。画面中有一条青色发光航线从左下或底部暗处出发，穿过画面中心，延伸到右上远方的港口灯光。航线不是旅游路线，而像一条被正式批准的行动路径。
+
+前景需要出现一张抽象“AI 船票”：不做真实纸票，不做复古票据，而是一张带金色边缘、青色数据线、登船编号感的半透明发光卡片。它可以位于画面右下或左下，与主标题错开，不遮挡文字。船票的作用是让“船票已经有了”这个隐喻成立。
+
+整体第一感觉应该是：正式开局、准备上船、路线已经亮起来。
+
+## Image Base Requirements
+
+生成图应该包含：
+
+- 16:9 横版全屏封面。
+- 深蓝黑商务底，带 AI 航线作战图质感。
+- 一条明确的青色发光航线，从暗处延伸到远处。
+- 远处港口 / 灯塔 / 航线终点的微弱暖金光。
+- 抽象 AI 船票，金色边缘 + 青色数据纹理。
+- 低调雷达扫描圈、罗盘刻度、坐标网格作为背景层。
+
+生成图不需要包含：
+
+- 具体人物。
+- 真实船只特写。
+- 会议室照片。
+- 复杂仪表盘 UI。
+- 可交互按钮、卡片堆、SaaS dashboard。
+
+允许少量文字直接进入图中。文字应尽量大、短、清晰。
+
+## Text / Typography Plan
+
+页面文字建议直接进入图中：
+
+主标题：
+
+> 搭上 AI 这条船
+
+副标题：
+
+> AI 智能化客户经营团队启动会
+
+角标小字：
+
+> 内部共识会议 / 启动培训前置会
+
+底部短句：
+
+> 先上船，再学怎么开船。
+
+标题层级：
+
+- 主标题为全页最大文字，放在画面中部偏左或中部偏上。
+- `AI` 可以用青色或淡金色强调，但不要做霓虹 cyberpunk。
+- 副标题放在主标题下方，字号明显小一档。
+- 角标小字放在左上或右上，像内部资料标记。
+- 底部短句放在底部安全区，作为收束金句。
+
+如果图像模型中文不够稳定，后续可接受小字轻微错漏；但主标题必须尽量清楚，不能被装饰压住。
+
+## Layout / Safe Areas
+
+建议构图：
+
+- 主标题区域：画面中部偏左，宽度约 55%-65%。
+- 航线主方向：从左下 / 下中出发，向右上远方延伸。
+- 船票：右下或左下前景，面积约画面 12%-18%，不能抢主标题。
+- 港口 / 灯塔远光：右上远方或画面深处，面积小，但要形成“目的地已亮”的感觉。
+- 雷达 / 罗盘图层：铺在背景，不要比标题更亮。
+- 四周保留边缘安全区，避免标题、角标、底句贴边。
+
+视觉重心顺序：
+
+1. 主标题；
+2. 发光航线；
+3. 抽象船票；
+4. 远方港口光；
+5. 雷达 / 罗盘 / 网格细节。
+
+## Color Guidance
+
+基础色：
+
+- 深蓝黑；
+- 暗海军蓝；
+- 冷灰蓝；
+- 近黑。
+
+信号色：
+
+- 青色用于航线、雷达、AI 信号；
+- 金色用于船票边缘、机会感、启动感；
+- 白色 / 淡蓝白用于标题。
+
+色彩边界：
+
+- 不要大面积紫色渐变。
+- 不要亮红金政府风。
+- 不要高饱和赛博霓虹。
+- 金色只能作为信号和重点，不要让画面变成奢华黑金海报。
+
+## Post-Production Notes
+
+本项目默认接受文字直接烘在图里。若 P01 图像整体气质很好，但文字轻微不稳，后续可以保留图像作为封面底图，再用 PPT 文字层覆盖主标题 / 副标题。此动作属于封面局部修正，不必回改公共视觉系统。
+
+如果出现以下情况，则属于公共视觉问题，需要回改 `visual-system/`：
+
+- 底板像 SaaS dashboard，而不是 AI 航线作战图；
+- 过度 cyberpunk；
+- 像政府汇报或客户方案；
+- 船票 / 航线隐喻不成立；
+- 深蓝黑底板压抑但没有开局感。
+
+## Constraints / Avoid
+
+必须避免：
+
+- 具体人物、群体、领导发言照；
+- 具象大船、邮轮、海盗船、童话船；
+- 可爱卡通船票；
+- 旅游广告海面；
+- 政府会议封面；
+- 客户提案封面；
+- 普通科技峰会海报；
+- 机器人头像、AI 大脑、芯片脸；
+- 文字太多、标题被挤压；
+- 大面积纯渐变背景，没有航线 / 雷达 / 罗盘逻辑。
+
+## Confirmation Gate
+
+请检查：
+
+1. 这个封面方向能否作为整套 PPT 的公共视觉样片；
+2. “AI 船票 / 正式上船 / 航线已亮”是否一眼成立；
+3. 主标题与副标题文案是否按现版本保留；
+4. 船票应放右下、左下，还是更靠中心。
+
+确认后再进入 `s01-v1-brief.md`、`s01-v1-prompt.md` 和封面图生成。
--- a/projects/2026-06-29-搭上AI这条船/slides/s01/s01-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s01/s01-v1-prompt.md
@ -0,0 +1,79 @@
+# S-01 Slide Prompt V1
+
+## Status
+
+- Status: ready for host-native generation.
+- Based on brief: `s01-v1-brief.md`
+- Review mode: none.
+- Garden/API expected: no.
+
+## Reference Inputs
+
+- Page: P01 / S-01.
+- Page family: A / Opening / Consensus.
+- Active style module: `visual-system/style/style-prompt-master.md`, Module A.
+- Core motifs: AI 航线, 船票, 港口 / 灯塔, 雷达.
+- Actual image references: none.
+- Reference note: no public visual reference image exists for this project yet; use the accepted text visual-system as the style authority.
+- Main title: `搭上 AI 这条船`
+- Core judgment: `船票已经有了，今天我们正式上船。`
+- Text complexity: low.
+- Generation path note: host-native / Advisor; Garden expected: no.
+
+## Operation Notes
+
+- `gpt-image-2` mode check result: `B-or-C`, recommendation `host-or-advisor`.
+- Skill/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Use built-in host image generation directly.
+- No child session, no review file, no completeness gate.
+
+## Advisor / Host-Native Record
+
+Mode:
+
+```json
+{
+  "mode": "B-or-C",
+  "recommendation": "host-or-advisor",
+  "garden_mode_enabled": false,
+  "has_api_key": false
+}
+```
+
+## Final Prompt
+
+Use case: productivity-visual
+Asset type: 16:9 full-page cover image for an internal PowerPoint kickoff meeting
+
+Create a 16:9 full-page visual slide image, cinematic executive business style. This is the cover page for an internal kickoff meeting titled "搭上 AI 这条船". The page should feel like a deep navy-black AI maritime operation map: a formal "boarding route" for a company team opening a new AI business line.
+
+Scene/backdrop: deep blue-black digital sea and maritime route-map background, subtle coordinate grid, compass markings, faint radar scan rings, restrained navigation-map texture. It should feel like an "AI航线作战图", not a software dashboard.
+
+Primary visual structure: a clear cyan glowing route line starts from a dark lower-left / lower-center origin and travels toward the upper-right distance. At the far destination, show small warm-gold port lights or a lighthouse-like signal, suggesting the route is now open. Add an abstract AI boarding ticket in the lower-right foreground: semi-transparent dark glass card, restrained gold edge, cyan data marks, no logo, no QR code focus. The ticket should be a supporting motif, not the main subject.
+
+Text (verbatim Chinese, keep large and readable):
+- Main title, largest: "搭上 AI 这条船"
+- Subtitle below: "AI 智能化客户经营团队启动会"
+- Bottom line: "先上船，再学怎么开船。"
+- Optional small corner label: "内部共识会议 / 启动培训前置会"
+
+Typography: bold modern Chinese sans-serif, high contrast, white or pale-blue title with very subtle cyan/gold accent on "AI". Keep text short and spacious. Main title must be the first read. Do not use decorative calligraphy.
+
+Composition/framing: title in the middle-left or mid-upper area, route line behind and around it without interfering with readability. Ticket in lower-right, about 12-18% of frame. Radar/compass/grid elements stay in the background. Keep safe margins around all text.
+
+Lighting/mood: energetic but disciplined, formal internal kickoff, "the route is ready, we are boarding now." Cinematic dark business atmosphere, not gloomy.
+
+Color palette: deep navy-black, dark slate blue, cyan AI signal light, restrained warm gold, pale-blue/white title.
+
+Constraints: no real people, no meeting room photo, no literal big ship, no cartoon ship, no pirate or fantasy ship, no humanoid robot, no AI face, no generic tech conference poster, no government-report red-gold style, no customer proposal style, no dense dashboard UI, no pure gradient background, no watermark, no logo.
+
+## Negative / Avoid
+
+Avoid unreadable small Chinese text, over-dense paragraphs, cyberpunk neon overload, luxury black-gold poster tone, cute illustration, stock-photo realism, cruise ship imagery, and UI cards.
+
+## Output Intent
+
+Generate `s01-v1-image.png` as the cover and public-style sample candidate.
--- a/projects/2026-06-29-搭上AI这条船/slides/s02/s02-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s02/s02-v1-brief.md
@ -0,0 +1,151 @@
+# S-02 Slide Brief V1
+
+## Status
+
+- Status: ready for prompt.
+- Source page: P02.
+- Output type: 16:9 full-page visual image.
+- Running mode: prompt and generate.
+- Review mode: none.
+- Write policy: append-only; overwrite allowed: no.
+
+## Source Basis
+
+- `../../intake/阶段5：P1-P24 单页设计骨架.md`
+- `../../visual-system/visual-system.md`
+- `../../visual-system/style/style-prompt-master.md`
+- `../../visual-system/motifs/motif-registry.md`
+- `s02-v1-design.md`
+
+Backup fact-check source recorded but not used:
+
+```text
+C:\Users\wangq\Documents\Codex\knowledge-vault\work\internal\OPC销售\AI智能化客户经营团队启动会议章程.md
+```
+
+## Slide Function
+
+明确会议性质，作为第一张内容页样片。它要说明：今天不是继续动员，而是把方案摊开形成共同方案。
+
+## First Read
+
+第一眼应该读到：
+
+> 今天这场会，不是再动员一次。
+
+第二眼读到左右对比：
+
+> 不是旧式动员；是共同定方案。
+
+第三眼读到右侧的行动节点：
+
+> 共识 / 补充 / 升级 / 训练。
+
+## Risk Class
+
+- Risk class: content-page readability / comparison infographic.
+- Public-style risk: medium, because this tests whether the common base supports normal content pages.
+- Page-specific risk: medium, because text density and left-right balance may need adjustment.
+
+## Reference Plan
+
+- Actual visual reference image: none.
+- Style source: `visual-system/style/style-prompt-master.md`
+- Motif source: `visual-system/motifs/motif-registry.md`
+- Active style module: Module A, Opening / Consensus.
+- Core motifs: AI 航线, route split, radar / operation-map grid, activated nodes.
+- Generation path: host-native image generation; Garden/API expected: no.
+
+## Visual Requirements
+
+- 16:9 horizontal full-page content slide image.
+- Same deep navy-black AI route operation-map base as P01.
+- Clear left-right comparison layout.
+- Left side: dark gray-blue, deactivated / closed route, marked `不是`.
+- Right side: active cyan/gold route, marked `是`.
+- Middle: thin glowing vertical route divider / checkpoint / routing gate.
+- Right side: four glowing action nodes: 共识, 补充, 升级, 训练.
+- Optional subtle top-down abstract meeting table linework in the background, very faint.
+
+## Character Handling
+
+No visible character. Do not add real people, meeting photos, speakers, or audience.
+
+## Composition / Scale
+
+- Header zone: top 15%-18%, title and core judgment.
+- Main comparison zone: middle 65%-70%.
+- Footer quote zone: bottom 10%-12%.
+- Left block: 42%-45%, low-light gray-blue.
+- Right block: 45%-48%, brighter cyan/gold.
+- Divider: 4%-6% center zone.
+
+## Light / Color / Texture
+
+- Base: same deep navy-black AI route operation-map base as S-01.
+- Left: muted slate gray / low-light blue-gray.
+- Right: active cyan route and restrained gold highlights.
+- Footer quote: small gold or pale-blue emphasis.
+- Texture: business operation map, not SaaS dashboard.
+
+## Text Plan
+
+Use in-image text. Minor text defects are acceptable, but the prompt must keep text compact.
+
+Required text priority:
+
+1. `今天这场会，不是再动员一次`
+2. `不是`
+3. `是`
+4. `共识`
+5. `补充`
+6. `升级`
+7. `训练`
+
+Secondary text if space and rendering stability allow:
+
+Left:
+
+- `再讲 AI 多重要`
+- `再劝要不要参与`
+- `我一个人宣布安排`
+- `听完就散`
+
+Right:
+
+- `摊开完整方案`
+- `补充不同角度`
+- `必要时升级方案`
+- `进入第一周训练`
+
+Footer:
+
+> 不是“我说你们听”，是“我们一起把这件事定下来”。
+
+If the generator cannot reliably handle all text, prioritize title, `不是 / 是`, and four right-side nodes. Secondary item text can be repaired later in PPT.
+
+## Forbidden Drift
+
+Avoid:
+
+- white consulting table;
+- ordinary dashboard UI;
+- real meeting-room photograph;
+- command-style policy page;
+- overly negative or accusatory `不是` side;
+- too much text;
+- separated style from S-01;
+- cyberpunk split line;
+- software project-management board.
+
+## Acceptance Checklist
+
+- The page remains visibly in the same deck style as S-01.
+- The `不是 / 是` contrast reads quickly.
+- The right-side path feels active and collaborative.
+- It feels like consensus formation, not orders from above.
+- Content is usable for an internal meeting even if minor text defects appear.
+
+## Prompt Handoff
+
+Create a host-native image prompt using Module A language and this brief. Do not use external reference images. Record `mode: B-or-C / host-or-advisor`. Generate directly in the main session after prompt creation, with no review file and no child session.
--- a/projects/2026-06-29-搭上AI这条船/slides/s02/s02-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s02/s02-v1-design.md
@ -0,0 +1,207 @@
+# S-02 Slide Design V1
+
+## Status
+
+- Status: draft for user review.
+- Source page: P02.
+- Page title: 今天这场会，不是再动员一次.
+- Role: first content-page sample.
+- Line: test/audit sample.
+- Write policy: append-only; overwrite allowed: no.
+- Next gate: user confirms design before brief / prompt / image generation.
+
+## Design Purpose
+
+本页用于验证：在 P01 的统一深蓝黑 AI 航线作战图底板上，普通内容页是否能读、能讲、能保持会议气质。
+
+它不是公共底板的重新设计，而是同一底板下的第一张内容页样片。它要解决的核心问题是：
+
+> 这场会不是继续动员，而是把方案摊开，形成共同方案。
+
+如果 P02 读不清、信息承载不稳、左右对比难成立，可能需要回改内容页规则；如果只是本页文案太多或左右布局比例不对，则只改 P02。
+
+## Discussion Summary
+
+P02 承接 P01 的封面，进入会议定位。用户希望内部会议 PPT 不追求精细排版，但每页必须形成判断、建立气氛。P02 作为内容页样片，应测试“同一视觉底板 + 一页一个判断 + 少量文字模型”的可行性。
+
+公共风格仍然不变：
+
+> 深蓝黑 AI 航线作战图底板，青色 / 金色高亮，雷达、航线、船票、罗盘等母题轻量变化。
+
+## Visual Concept
+
+画面是同一张 AI 航线作战图底板上的“会议定位分岔图”。
+
+顶部放主标题：
+
+> 今天这场会，不是再动员一次
+
+画面主体是左右对比：
+
+- 左侧是暗灰、降噪、低亮度的“不是”区域，像旧航线 / 无效循环 / 已关闭任务；
+- 右侧是青色与金色高亮的“是”区域，像当前被激活的行动路线；
+- 中间不是普通竖线，而是一条类似航线分界 / 闸门 / 作战地图分割线的发光细线，让它仍然属于统一视觉系统。
+
+右侧可以叠加四个发光节点：共识、补充、升级、训练。它们是本页的行动路径。
+
+整体第一感觉应该是：今天不是喊口号，而是进入正式协作。
+
+## Image Base Requirements
+
+生成图应该包含：
+
+- 16:9 横版内容页。
+- 与 P01 同一底板气质：深蓝黑、AI 航线作战图、雷达 / 坐标 / 路线。
+- 左右对比结构。
+- 左侧“不是”低亮度灰色，像被关闭的旧路线。
+- 右侧“是”青色 / 金色高亮，像被激活的新路线。
+- 中间有一条发光分界线或航线闸门。
+- 右侧有“共识 / 补充 / 升级 / 训练”四个发光节点。
+- 背景可以有抽象会议桌俯视线框，但必须很淡，不能变成会议室照片。
+
+生成图不需要包含：
+
+- 真实参会人物；
+- 会议桌照片；
+- 卡通插画；
+- 传统 PPT 白底左右表格；
+- 笨重 UI 卡片堆。
+
+## Text / Typography Plan
+
+页面文字建议直接进入图中，但要控制长度。
+
+主标题：
+
+> 今天这场会，不是再动员一次
+
+核心判断：
+
+> 大家已经有意愿，今天要形成共同方案。
+
+左侧标题：
+
+> 不是
+
+左侧 4 个短项：
+
+- 再讲 AI 多重要
+- 再劝要不要参与
+- 我一个人宣布安排
+- 听完就散
+
+右侧标题：
+
+> 是
+
+右侧 4 个短项：
+
+- 摊开完整方案
+- 补充不同角度
+- 必要时升级方案
+- 进入第一周训练
+
+底部金句：
+
+> 不是“我说你们听”，是“我们一起把这件事定下来”。
+
+文字优先级：
+
+1. 主标题；
+2. 左右大字“不是 / 是”；
+3. 右侧四个行动节点；
+4. 核心判断；
+5. 底部金句；
+6. 左右细项。
+
+如果图像模型难以稳定生成全部文字，优先保留：主标题、不是、是、共识、补充、升级、训练。细项可在后续 PPT 层补充或压缩。
+
+## Layout / Safe Areas
+
+建议构图：
+
+- 顶部 15%-18% 放主标题和一句核心判断。
+- 主体 65%-70% 做左右对比。
+- 底部 10%-12% 放金句。
+- 左侧占 42%-45%，低亮度灰蓝。
+- 右侧占 45%-48%，青色 / 金色更亮。
+- 中间留 4%-6% 给发光分界线、航线闸门或路由分叉。
+
+左侧视觉：
+
+- 灰色旧任务列表；
+- 轻微斜线 / 断开的路线；
+- 不要出现失败、否定、崩塌的强烈灾难感。
+
+右侧视觉：
+
+- 四个发光节点按从上到下或从左下到右上的路径排列；
+- 节点文字为：共识、补充、升级、训练；
+- 路线最终指向“第一周训练”。
+
+安全区：
+
+- 左右文字不要贴边；
+- 底部金句不要靠近最下缘；
+- 中间分界线不要穿过标题；
+- 背景雷达线不能干扰正文。
+
+## Color Guidance
+
+整体延续 P01：
+
+- 深蓝黑底；
+- 冷灰蓝结构线；
+- 青色路线和节点；
+- 金色用于“是”侧的关键节点和底部金句；
+- 左侧“不是”使用灰色 / 低饱和蓝灰。
+
+对比关系：
+
+- 左侧不是要“死黑”，而是“已关闭 / 不再走”的低亮状态；
+- 右侧是当前激活路线，要更有能量；
+- 不要把左侧做得太负面，会议气质应是共识式，不是批判式。
+
+## Post-Production Notes
+
+本页是内容页样片，允许出现轻微文字错漏。但如果左右结构和主判断成立，可以通过后续 PPT 文字层修正细项。
+
+如果出现以下情况，属于公共 / 内容系统问题，需要回改 `visual-system` 或 `style-prompt-master`：
+
+- 同一底板无法承载内容页，文字区域被背景干扰严重；
+- 内容页变成普通 UI dashboard；
+- 左右对比页与封面风格断裂；
+- “共识式会议”气质消失，变成命令式制度页。
+
+如果出现以下情况，只改 P02：
+
+- 左右比例不舒服；
+- 细项文字太多；
+- 底部金句位置压迫；
+- “不是 / 是”的亮度对比不够。
+
+## Constraints / Avoid
+
+必须避免：
+
+- 白底咨询公司左右表格；
+- 大量卡片堆叠；
+- 真实会议室照片；
+- 领导讲话台；
+- 命令式制度宣贯；
+- 错把“不是”做成攻击或批判；
+- 右侧节点像项目管理软件；
+- 文字密度过高导致全页不可读；
+- 过亮 cyberpunk 分割线破坏商务感。
+
+## Confirmation Gate
+
+请检查：
+
+1. P02 是否能作为内容页样片；
+2. “不是 / 是”的左右对比是否符合你的会议气质；
+3. 右侧是否保留“共识 / 补充 / 升级 / 训练”四个节点；
+4. 底部金句是否保留，还是压缩为更短版本；
+5. 如果 P01 通过后，P02 是否可以直接进入 brief / prompt / 生成。
+
+确认后再进入 `s02-v1-brief.md`、`s02-v1-prompt.md` 和内容页样片生成。
--- a/projects/2026-06-29-搭上AI这条船/slides/s02/s02-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s02/s02-v1-prompt.md
@ -0,0 +1,86 @@
+# S-02 Slide Prompt V1
+
+## Status
+
+- Status: ready for host-native generation.
+- Based on brief: `s02-v1-brief.md`
+- Review mode: none.
+- Garden/API expected: no.
+
+## Reference Inputs
+
+- Page: P02 / S-02.
+- Page family: A / Opening / Consensus.
+- Active style module: `visual-system/style/style-prompt-master.md`, Module A.
+- Core motifs: AI 航线, route split, activated nodes, operation-map grid.
+- Actual image references: none.
+- Reference note: no public visual reference image exists for this project yet; use the accepted text visual-system as the style authority.
+- Main title: `今天这场会，不是再动员一次`
+- Core judgment: `大家已经有意愿，今天要形成共同方案。`
+- Text complexity: medium.
+- Generation path note: host-native / Advisor; Garden expected: no.
+
+## Operation Notes
+
+- `gpt-image-2` mode check result: `B-or-C`, recommendation `host-or-advisor`.
+- Skill/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Use built-in host image generation directly.
+- No child session, no review file, no completeness gate.
+
+## Advisor / Host-Native Record
+
+Mode:
+
+```json
+{
+  "mode": "B-or-C",
+  "recommendation": "host-or-advisor",
+  "garden_mode_enabled": false,
+  "has_api_key": false
+}
+```
+
+## Final Prompt
+
+Use case: productivity-visual
+Asset type: 16:9 full-page content slide image for an internal PowerPoint kickoff meeting
+
+Create a 16:9 full-page visual slide image in the same unified style as a deep navy-black AI maritime operation map. This is the first content page after the cover. It should explain meeting positioning: this meeting is not another mobilization; it is a collaborative session to form a shared plan.
+
+Scene/backdrop: deep blue-black executive AI route-map base, subtle coordinate grid, radar scan rings, compass/navigation marks, faint abstract top-down meeting-table linework as background only. Keep the same visual base as the cover: disciplined internal AI route / war-map atmosphere, not a white consulting slide and not a SaaS dashboard.
+
+Main layout: clear left-right comparison structure.
+- Left side: low-light gray-blue deactivated route area labeled "不是". It should feel like a closed old route or inactive task list, not a harsh criticism.
+- Right side: active cyan and restrained gold route area labeled "是". It should feel like the chosen active route.
+- Center: a thin glowing vertical route divider / checkpoint / routing gate, integrated into the maritime operation map.
+- Right side should include four glowing action nodes connected by a subtle route path: "共识", "补充", "升级", "训练".
+
+Text (verbatim Chinese, prioritize readability):
+- Main title at top, large: "今天这场会，不是再动员一次"
+- Core judgment near title, smaller: "大家已经有意愿，今天要形成共同方案"
+- Left heading: "不是"
+- Left short items: "再讲 AI 多重要", "再劝要不要参与", "我一个人宣布安排", "听完就散"
+- Right heading: "是"
+- Right short items: "摊开完整方案", "补充不同角度", "必要时升级方案", "进入第一周训练"
+- Footer quote, optional if space allows: "不是“我说你们听”，是“我们一起把这件事定下来”"
+
+Typography: modern Chinese sans-serif, high contrast, large title, clear headings. Use short labels and avoid tiny text. If not all text can fit cleanly, prioritize the main title, the headings "不是" and "是", and the four nodes "共识 / 补充 / 升级 / 训练".
+
+Composition/framing: top 15-18% for title and core judgment, middle 65-70% for comparison, bottom 10-12% for footer quote. Left block about 43% width, right block about 47% width, center divider about 5%. Keep safe margins, avoid text touching edges, and do not let background radar lines interfere with text.
+
+Lighting/mood: consensus-building, pragmatic, energetic but not loud. The right side should feel active and hopeful; the left side muted but not hostile.
+
+Color palette: same deep navy-black base, muted slate gray-blue on the left, active cyan route and restrained warm gold highlights on the right, pale-blue/white text.
+
+Constraints: no real people, no meeting-room photo, no leader podium, no government policy page, no command-style discipline board, no ordinary dashboard UI, no white consulting table, no excessive cards, no cyberpunk neon split line, no cartoon illustration, no humanoid robots, no logo, no watermark.
+
+## Negative / Avoid
+
+Avoid making the page look like a project-management app, a customer proposal, a government report, or a dense training manual. Avoid making the "不是" side feel accusatory or negative beyond a muted inactive route.
+
+## Output Intent
+
+Generate `s02-v1-image.png` as the first content-page sample candidate.
--- a/projects/2026-06-29-搭上AI这条船/slides/s03/s03-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s03/s03-v1-brief.md
@ -0,0 +1,12 @@
+# s03-v1 Brief
+
+- Slide: s03 / P03
+- Title: 我们要建的，不是传统销售团队
+- Core judgment: 我们要建的是 AI 智能化客户经营团队。
+- Visual model: radar-centered team identity map.
+- Required text: title, central team name, five short capability labels, one bottom sentence.
+- Style base: accepted deep navy-black AI route operation map with cyan radar signals and gold emphasis.
+- Motifs: radar, route map, collaboration nodes.
+- Avoid: traditional sales proposal look, customer-facing SaaS dashboard, cartoon ship, humanoid robots, concrete attendee portraits.
+- Output target: `projects/2026-06-29-搭上AI这条船/slides/s03/s03-v1-image.png`
+- Review mode: none.
--- a/projects/2026-06-29-搭上AI这条船/slides/s03/s03-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s03/s03-v1-design.md
@ -0,0 +1,28 @@
+# s03-v1 Design
+
+## 页面
+
+P03：我们要建的，不是传统销售团队
+
+## 设计目标
+
+把团队身份从“传统销售”切换成“AI 智能化客户经营团队”。画面要让观众一眼看到：中心不是卖货岗位，而是一个在雷达中心识别客户、经营关系、管理预期、发现机会、推动合作的新型前端团队。
+
+## 画面概念
+
+使用统一深蓝黑 AI 航线作战图底板。中心是发光雷达盘，核心文字为“AI 智能化客户经营团队”。外圈五个能力点围成雷达环：识别客户、经营关系、管理预期、发现机会、推动合作。最外圈用四个较小协作节点表示客户、专家、AI 业务工程师、工程团队。
+
+## 文案布局
+
+- 顶部大标题：我们要建的，不是传统销售团队
+- 中央主词：AI 智能化客户经营团队
+- 外圈短标签：识别客户 / 经营关系 / 管理预期 / 发现机会 / 推动合作
+- 底部短句：不是卖课、卖软件、卖项目，而是经营客户智能化旅程
+
+## 视觉要求
+
+延续 s01/s02 的深海航线地图、青色雷达线、金色重点标签、白色大标题。页面保持信息图感，不画真实人物，不做 SaaS 仪表盘，不做可爱船插画。
+
+## 后期提示
+
+如果生成文字有轻微瑕疵，可在 PPT 中覆盖标题和底部短句；核心视觉必须保留“雷达中心团队身份”。
--- a/projects/2026-06-29-搭上AI这条船/slides/s03/s03-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s03/s03-v1-prompt.md
@ -0,0 +1,40 @@
+# s03-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used:
+  - `AGENTS.md`
+  - `projects/2026-06-29-搭上AI这条船/handoff/2026-06-29-ai-ship-formal-production-handoff.md`
+  - `projects/2026-06-29-搭上AI这条船/project.md`
+  - `projects/2026-06-29-搭上AI这条船/execution-plan.md`
+  - `projects/2026-06-29-搭上AI这条船/slides/slides.md`
+  - `projects/2026-06-29-搭上AI这条船/visual-system/visual-system.md`
+  - `projects/2026-06-29-搭上AI这条船/visual-system/style/style-prompt-master.md`
+  - `projects/2026-06-29-搭上AI这条船/visual-system/motifs/motif-registry.md`
+  - `projects/2026-06-29-搭上AI这条船/intake/阶段5：P1-P24 单页设计骨架.md`
+  - `projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/task-plan.md`
+  - `projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/production-run-status.md`
+- Visual calibration references loaded:
+  - Image 1: accepted cover style sample, `slides/s01/s01-v1-image.png`
+  - Image 2: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual, matching the accepted samples: deep navy-black AI route operation map, subtle ocean/chart texture, thin grid lines, compass/radar overlays, cyan route glow, gold accent points, crisp white Chinese title typography. Use Image 1 for the overall deep route-map atmosphere and Image 2 for the content-page layout density and text hierarchy.
+
+Page title at top: “我们要建的，不是传统销售团队”.
+
+Centerpiece: a large glowing cyan radar disk in the middle of the slide. In the radar center, place the main phrase in bold readable Chinese: “AI 智能化客户经营团队”. Around the radar ring, place five short capability labels as compact chips or orbit nodes: “识别客户”, “经营关系”, “管理预期”, “发现机会”, “推动合作”. On the outer ring, add four smaller collaboration nodes labeled “客户”, “专家”, “AI业务工程师”, “工程团队”, connected by fine cyan lines.
+
+Bottom sentence, smaller but readable: “不是卖课、卖软件、卖项目，而是经营客户智能化旅程”.
+
+Visual mood: confident, operational, internal route-map briefing. Keep it elegant and cinematic, not a government report, not a customer proposal, not a SaaS dashboard. No humanoid robots, no cartoon ship, no real people or portraits. Use only one title, one central judgment, short labels, and clean information hierarchy. Minor Chinese text imperfections are acceptable, but make the text as legible as possible.
--- a/projects/2026-06-29-搭上AI这条船/slides/s04/s04-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s04/s04-v1-brief.md
@ -0,0 +1,12 @@
+# s04-v1 Brief
+
+- Slide: s04 / P04
+- Title: 为什么现在必须变
+- Core judgment: 旧船越来越难跑，新船已经靠岸。
+- Visual model: old route pressure versus new AI route.
+- Required text: title, three short labels, one bottom judgment.
+- Style base: accepted deep navy-black AI route operation map with cyan/gold signals.
+- Motifs: old dock, new route, lighthouse, turning point.
+- Avoid: macroeconomic report, gloomy disaster page, customer proposal, cyberpunk poster, cartoon ship.
+- Output target: `projects/2026-06-29-搭上AI这条船/slides/s04/s04-v1-image.png`
+- Review mode: none.
--- a/projects/2026-06-29-搭上AI这条船/slides/s04/s04-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s04/s04-v1-design.md
@ -0,0 +1,29 @@
+# s04-v1 Design
+
+## 页面
+
+P04：为什么现在必须变
+
+## 设计目标
+
+建立现实压力，但不制造悲观。画面要表达“旧船越来越难跑，新船已经靠岸”：左侧旧行业航线变暗、下降，右侧 AI 新航线被点亮，中间出现必须转向的节点。
+
+## 画面概念
+
+统一深蓝黑航线作战图底板。左右对比结构：左侧是暗色旧码头、下降曲线、灰蓝航道；右侧是亮起的 AI 航线、远方灯塔、金色靠岸信号。中间用一个“转向点 / 航向切换”符号连接两侧。
+
+## 文案布局
+
+- 顶部大标题：为什么现在必须变
+- 左侧短句：旧行业下滑
+- 中间短句：转向点
+- 右侧短句：AI 新入口
+- 底部金句：旧船越来越难跑，新船已经靠岸
+
+## 视觉要求
+
+整体有压力与转机，不做宏观趋势报告。左侧低亮度但不脏，右侧青色路径和金色灯塔作为希望点。保留航线图、雷达和指南针语言。
+
+## 后期提示
+
+底部金句是本页核心；若图中文字不稳，优先覆盖底部金句和右侧“AI 新入口”。
--- a/projects/2026-06-29-搭上AI这条船/slides/s04/s04-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s04/s04-v1-prompt.md
@ -0,0 +1,29 @@
+# s04-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P04 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted cover style sample, `slides/s01/s01-v1-image.png`
+  - Image 2: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual in the same style as the accepted samples: deep navy-black maritime AI operation chart, subtle ocean route map, compass grid, radar sweeps, cyan route glow, restrained gold emphasis, large crisp white Chinese typography.
+
+Page title at top: “为什么现在必须变”.
+
+Layout: left-right contrast with a central turning point. Left side: dim old dock / old route zone, a muted downward line, low blue-gray signal, label “旧行业下滑”. Right side: bright cyan AI route opening toward a warm gold lighthouse, label “AI 新入口”. Center: a clear navigation pivot marker labeled “转向点”, with route line bending from old path to new path.
+
+Bottom core judgment in one strong line: “旧船越来越难跑，新船已经靠岸”.
+
+Tone: pressure plus opportunity, grounded in real business anxiety but not pessimistic. It should feel like an internal route-map briefing, not a macro trend report. Avoid government report style, customer proposal style, SaaS dashboard UI, cyberpunk poster, cute ship cartoon, robots, and real people. Keep text short and legible.
--- a/projects/2026-06-29-搭上AI这条船/slides/s05/s05-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s05/s05-v1-brief.md
@ -0,0 +1,12 @@
+# s05-v1 Brief
+
+- Slide: s05 / P05
+- Title: 我这半年打通了什么
+- Core judgment: 产品力、方法论、训练路径，已经准备到可以开局。
+- Visual model: ship deck capability base with three pillars.
+- Required text: title, three labels, one bottom sentence.
+- Style base: deep navy-black AI route operation map, cyan/gold signal hierarchy.
+- Motifs: route map, ship deck structure, capability pillars.
+- Avoid: personal resume, trophy wall, customer proposal, cartoon ship, humanoid robots.
+- Output target: `projects/2026-06-29-搭上AI这条船/slides/s05/s05-v1-image.png`
+- Review mode: none.
--- a/projects/2026-06-29-搭上AI这条船/slides/s05/s05-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s05/s05-v1-design.md
@ -0,0 +1,31 @@
+# s05-v1 Design
+
+## 页面
+
+P05：我这半年打通了什么
+
+## 设计目标
+
+建立“强哥已经准备好带大家开局”的可信度，但避免个人履历炫耀。画面表达三项底座已经成型：产品力、方法论、训练路径。
+
+## 画面概念
+
+统一深蓝黑航线作战图底板。底部是一条抽象船体 / 甲板结构图，不画卡通船。甲板上竖起三根发光桅杆或三座能力塔，分别标注“产品力”“方法论”“训练路径”。三根桅杆向上连接到同一条 AI 航线，表示已经可以开局。
+
+## 文案布局
+
+- 顶部大标题：我这半年打通了什么
+- 三列主标签：产品力 / 方法论 / 训练路径
+- 每列一句短解释：
+  - 三级路径
+  - 判断框架
+  - 销售老兵训练法
+- 底部短句：路径终于想清楚了
+
+## 视觉要求
+
+金色用于“已经准备好”的可信节点，青色用于连接航线和结构线。整页要像能力底座图，不像个人荣誉墙。
+
+## 后期提示
+
+保留三列结构即可；详细解释可在 PPT 层补字。
--- a/projects/2026-06-29-搭上AI这条船/slides/s05/s05-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s05/s05-v1-prompt.md
@ -0,0 +1,29 @@
+# s05-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P05 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted cover style sample, `slides/s01/s01-v1-image.png`
+  - Image 2: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT visual. Match the accepted samples: deep navy-black AI maritime operation chart, subtle ocean map texture, compass/radar overlays, cyan glowing route lines, restrained gold accents, crisp white Chinese type, cinematic but businesslike.
+
+Page title at top: “我这半年打通了什么”.
+
+Main visual: an abstract ship deck / structural base diagram along the lower third, not a cartoon ship. On the deck stand three luminous capability pillars or mast-like towers, evenly spaced, connected upward into one cyan AI route line. Label the three pillars clearly: “产品力”, “方法论”, “训练路径”. Under each, use one short small label: “三级路径”, “判断框架”, “训练方法”.
+
+Bottom sentence: “路径终于想清楚了，可以开局”.
+
+Make the page feel like readiness and structured preparation, not personal bragging. Use gold highlights for readiness nodes and cyan for route/connection lines. Avoid resume-like visuals, trophy imagery, government report style, SaaS dashboard UI, cute ship cartoon, robots, and real people. Keep text short and legible.
--- a/projects/2026-06-29-搭上AI这条船/slides/s06/s06-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s06/s06-v1-brief.md
@ -0,0 +1,12 @@
+# s06-v1 Brief
+
+- Slide: s06 / P06
+- Title: 我的角色已经变了
+- Core judgment: 我不能再长期同时承担所有角色。
+- Visual model: overloaded center node to expert network node.
+- Required text: title, left old role, right new role, four short expert functions, bottom sentence.
+- Style base: deep navy-black AI route operation map with cyan/gold signals.
+- Motifs: pressure node, network map, role transition arrow.
+- Avoid: real portrait, boss biography, blame-shifting tone, customer proposal, SaaS dashboard, robots.
+- Output target: `projects/2026-06-29-搭上AI这条船/slides/s06/s06-v1-image.png`
+- Review mode: none.
--- a/projects/2026-06-29-搭上AI这条船/slides/s06/s06-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s06/s06-v1-design.md
@ -0,0 +1,29 @@
+# s06-v1 Design
+
+## 页面
+
+P06：我的角色已经变了
+
+## 设计目标
+
+说明强哥角色从“全能型老板”转向“学者型实践专家 / 方法论与判断节点”。观众应理解这是体系升级，不是退出前线或甩锅。
+
+## 画面概念
+
+统一深蓝黑作战图底板。左右转型结构：左侧是一个高压中心节点，周围堆叠多个旧标签；右侧是一个更稳的专家节点，连接客户经营团队、AI 业务工程师、工程底座等多个节点。中间用青色箭头表示角色换位。
+
+## 文案布局
+
+- 顶部大标题：我的角色已经变了
+- 左侧标签：全能型老板
+- 右侧标签：学者型实践专家
+- 右侧短标签：方法论 / 认知引导 / 场景判断 / 质量把关
+- 底部短句：我不是退出前线，而是换到更该站的位置
+
+## 视觉要求
+
+不使用人物肖像。用节点、网络、压力环和航线表达角色变化。左侧压力感，右侧秩序感。
+
+## 后期提示
+
+如果需要 PPT 层补充旧角色长列表，放在左侧灰色标签区；生成图只保留短标签。
--- a/projects/2026-06-29-搭上AI这条船/slides/s06/s06-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s06/s06-v1-prompt.md
@ -0,0 +1,29 @@
+# s06-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P06 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted cover style sample, `slides/s01/s01-v1-image.png`
+  - Image 2: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual, matching the accepted samples: deep navy-black AI route operation map, compass grid, radar sweep lines, cyan route glow, gold emphasis, crisp white Chinese typography.
+
+Page title at top: “我的角色已经变了”.
+
+Layout: left-to-right transformation. Left side: a single overloaded glowing node under pressure rings, labeled “全能型老板”, surrounded by small muted tags like “专家”, “商务”, “销售”, “方案”, “谈判”, “把关”. Right side: a stable expert node connected to several team nodes, labeled “学者型实践专家”. Around it place four short function labels: “方法论”, “认知引导”, “场景判断”, “质量把关”. Center: a clean cyan transition arrow or route switch from left to right.
+
+Bottom sentence: “我不是退出前线，而是换到更该站的位置”.
+
+Mood: system upgrade, calm authority, clear role repositioning. No human portraits, no realistic boss figure, no blame tone. Avoid customer proposal style, SaaS dashboard UI, government report, robots, cyberpunk poster, cartoon ship. Keep all Chinese text short and as legible as possible.
--- a/projects/2026-06-29-搭上AI这条船/slides/s07/s07-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s07/s07-v1-brief.md
@ -0,0 +1,12 @@
+# s07-v1 Brief
+
+- Slide: s07 / P07
+- Title: 过去靠我一个人顶，未来靠体系一起跑
+- Core judgment: 个人驱动必须升级为团队驱动。
+- Visual model: single engine to multi-engine system.
+- Required text: title, past/future labels, four future keywords, bottom judgment.
+- Style base: deep navy-black AI route operation map, cyan/gold signal hierarchy.
+- Motifs: route map, single engine, multi-node fleet, distributed propulsion.
+- Avoid: complaint tone, cute ship, real people, customer proposal, SaaS dashboard, robots.
+- Output target: `projects/2026-06-29-搭上AI这条船/slides/s07/s07-v1-image.png`
+- Review mode: none.
--- a/projects/2026-06-29-搭上AI这条船/slides/s07/s07-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s07/s07-v1-design.md
@ -0,0 +1,29 @@
+# s07-v1 Design
+
+## 页面
+
+P07：过去靠我一个人顶，未来靠体系一起跑
+
+## 设计目标
+
+从个人角色变化自然过渡到团队建设。画面要表达个人驱动升级为团队驱动：过去一个中心发动机承压，未来多节点协同推进。
+
+## 画面概念
+
+统一深蓝黑航线作战图底板。上下结构：上半部分是单引擎小船 / 单中心推进节点，周围有风险和压力标记；下半部分是多引擎船队或多节点协同网络，几条青色航线一起向前。整体像作战路线升级图，不画可爱船。
+
+## 文案布局
+
+- 顶部大标题：过去靠我一个人顶，未来靠体系一起跑
+- 上半标签：过去：个人驱动
+- 下半标签：未来：团队驱动
+- 下方关键词：团队分工 / AI 工具 / 客户经营 / 流程机制
+- 底部金句：个人驱动必须升级为团队驱动
+
+## 视觉要求
+
+上半压力低亮，下半协同高亮。表达“体系才能跑得远”，不是抱怨个人太累。
+
+## 后期提示
+
+长列表不进入生成图；必要时在 PPT 旁白讲。
--- a/projects/2026-06-29-搭上AI这条船/slides/s07/s07-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s07/s07-v1-prompt.md
@ -0,0 +1,29 @@
+# s07-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P07 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted cover style sample, `slides/s01/s01-v1-image.png`
+  - Image 2: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT visual. Match the accepted samples: deep navy-black AI route operation chart, subtle maritime map texture, compass/radar overlays, cyan glowing route paths, gold key nodes, large white Chinese title, crisp diagram labels.
+
+Page title at top: “过去靠我一个人顶，未来靠体系一起跑”.
+
+Layout: upper half and lower half. Upper half: a single overloaded propulsion node / small abstract vessel icon, muted blue-gray, labeled “过去：个人驱动”, with pressure rings and several small dim tags. Lower half: a brighter distributed system, multiple engine nodes / small fleet route nodes connected by cyan lines, labeled “未来：团队驱动”. Add four compact labels near the future network: “团队分工”, “AI工具”, “客户经营”, “流程机制”.
+
+Bottom core judgment: “个人驱动必须升级为团队驱动”.
+
+Mood: organizational upgrade, confidence, not exhaustion or complaint. It should feel like an internal operation map. Avoid customer proposal style, SaaS dashboard, government report, cute cartoon ship, humanoid robots, real people. Keep text short and legible.
--- a/projects/2026-06-29-搭上AI这条船/slides/s08/s08-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s08/s08-v1-brief.md
@ -0,0 +1,12 @@
+# s08-v1 Brief
+
+- Slide: s08 / P08
+- Title: 为什么不是传统销售
+- Core judgment: 这件事不是把 AI 卖出去，而是把客户带上正确路径。
+- Visual model: traditional sales funnel versus AI customer journey route.
+- Required text: title, two side headers, short labels, bottom sentence.
+- Style base: deep navy-black AI route operation map with cyan/gold signals.
+- Motifs: funnel, route line, lighthouse, boundary markers.
+- Avoid: insulting traditional sales, SaaS dashboard, customer proposal, cartoon ship, robots, portraits.
+- Output target: `projects/2026-06-29-搭上AI这条船/slides/s08/s08-v1-image.png`
+- Review mode: none.
--- a/projects/2026-06-29-搭上AI这条船/slides/s08/s08-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s08/s08-v1-design.md
@ -0,0 +1,28 @@
+# s08-v1 Design
+
+## 页面
+
+P08：为什么不是传统销售
+
+## 设计目标
+
+重定义销售工作：传统销售能力仍然有用，但 AI 项目要升级为客户经营和路径推进。画面不能贬低传统销售，而是表达“发动机换了”。
+
+## 画面概念
+
+统一深蓝黑航线作战图底板。左右对比：左侧是传统成交漏斗，短标签为找客户、讲产品、谈价格、促成交、追回款；右侧是客户旅程航线，短标签为看懂客户、判断基础、识别场景、管理预期、推进路径、保护边界。右侧航线更亮，连接到远方 AI 灯塔。
+
+## 文案布局
+
+- 顶部大标题：为什么不是传统销售
+- 左侧标题：传统销售
+- 右侧标题：AI 客户经营
+- 底部短句：不是卖东西，是经营客户的智能化旅程
+
+## 视觉要求
+
+左侧不要做成失败或落后，只是“旧打法”；右侧体现升级后的路径感。青色用于新旅程，金色用于边界与关键判断。
+
+## 后期提示
+
+如果生成图左右文字不够稳定，可在 PPT 中重做标签；关键是漏斗 vs 航线结构。
--- a/projects/2026-06-29-搭上AI这条船/slides/s08/s08-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s08/s08-v1-prompt.md
@ -0,0 +1,29 @@
+# s08-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P08 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted cover style sample, `slides/s01/s01-v1-image.png`
+  - Image 2: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual, same accepted style: deep navy-black AI operation route map, subtle ocean chart texture, compass/radar elements, cyan route glow, gold accent nodes, crisp white Chinese headings.
+
+Page title at top: “为什么不是传统销售”.
+
+Layout: respectful left-right comparison. Left side: a muted but clean sales funnel labeled “传统销售”, with short stacked labels: “找客户”, “讲产品”, “谈价格”, “促成交”, “追回款”. Right side: a brighter customer journey route map labeled “AI 客户经营”, with route nodes labeled: “看懂客户”, “判断基础”, “识别场景”, “管理预期”, “推进路径”, “保护边界”. The right route leads toward a small gold lighthouse / AI route marker.
+
+Bottom sentence: “不是卖东西，是经营客户的智能化旅程”.
+
+The page must say “upgrade”, not “old sales are useless”. Keep the left side dignified, right side more future-facing. Avoid customer proposal style, SaaS dashboard UI, government report, cartoon ship, robots, real people. Use short readable Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s09/s09-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s09/s09-v1-brief.md
@ -0,0 +1,12 @@
+# s09-v1 Brief
+
+- Slide: s09 / P09
+- Title: 为什么这次找你们
+- Core judgment: AI 客户经营第一阶段，最重要的是识人、识局、识势。
+- Visual model: radar-scanned customer organization plus three-ability triangle.
+- Required text: title, three large words, short labels, bottom sentence.
+- Style base: deep navy-black AI route operation map with cyan/gold signals.
+- Motifs: radar, customer organization nodes, boarding opportunity signal.
+- Avoid: training lecture page, portraits, cartoon ship, robots, customer proposal, SaaS dashboard.
+- Output target: `projects/2026-06-29-搭上AI这条船/slides/s09/s09-v1-image.png`
+- Review mode: none.
--- a/projects/2026-06-29-搭上AI这条船/slides/s09/s09-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s09/s09-v1-design.md
@ -0,0 +1,31 @@
+# s09-v1 Design
+
+## 页面
+
+P09：为什么这次找你们
+
+## 设计目标
+
+建立参会者信心。观众要感到自己被需要，不是因为懂技术，而是因为懂人、懂局、懂客户。核心三词：识人、识局、识势。
+
+## 画面概念
+
+统一深蓝黑 AI 航线作战图底板。中心是客户组织雷达扫描盘，雷达上有不同颜色的小节点表示决策者、推动者、阻力者、旁观者。三角模型托起三项能力：识人、识局、识势。底座写“多年客户经验 / 社会经验 / 谈判经验”。
+
+## 文案布局
+
+- 顶部大标题：为什么这次找你们
+- 三个大词：识人 / 识局 / 识势
+- 每个词一句短解释：
+  - 真决策者
+  - 推动与阻力
+  - 真需求与看热闹
+- 底部金句：老销售的能力没有过时，只是换了发动机
+
+## 视觉要求
+
+这页要让人抬头，有尊重感。不要做培训说教页，不要把人画成头像。用雷达和组织节点表示经验的价值。
+
+## 后期提示
+
+如文字不稳，优先保留三个大词和底部金句。
--- a/projects/2026-06-29-搭上AI这条船/slides/s09/s09-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s09/s09-v1-prompt.md
@ -0,0 +1,29 @@
+# s09-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P09 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted cover style sample, `slides/s01/s01-v1-image.png`
+  - Image 2: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual. Match the accepted samples: deep navy-black AI maritime route operation map, compass/radar grid, subtle ocean chart, cyan signal lines, gold highlights, large readable white Chinese title.
+
+Page title at top: “为什么这次找你们”.
+
+Main visual: a customer-organization radar scan. Center or middle-right contains a glowing radar disk with small colored nodes representing roles in a customer organization: “决策者”, “推动者”, “阻力者”, “旁观者”. Over or beside the radar, build a clear three-point triangle model with three large gold/cyan words: “识人”, “识局”, “识势”. Under each word add a very short label: “真决策者”, “推动与阻力”, “真需求”.
+
+Bottom sentence: “老销售的能力没有过时，只是换了发动机”.
+
+Mood: respect, confidence, invitation into the AI era. It should make experienced salespeople feel newly valuable. Avoid training lecture look, government report, customer proposal, SaaS dashboard, cute ship cartoon, robots, real portraits or attendee faces. Use short, legible Chinese labels; one title, one core judgment, no long paragraphs.
--- a/projects/2026-06-29-搭上AI这条船/slides/s10/s10-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s10/s10-v1-brief.md
@ -0,0 +1,32 @@
+# s10-v1 Brief
+
+## Page
+
+- Unit: s10
+- Source page: P10
+- Title: Agentic 技术平权：外行也能上船
+- Page family: Confidence / Agentic Entry
+- Version: v1
+
+## Core Conclusion
+
+技术平权，不是人人都当程序员，而是外行也能参与探索。
+
+## Objective
+
+让参会者相信，AI Agentic 阶段的第一步不是写代码，而是用业务理解、自然语言和任务拆解进入探索轨。
+
+## Visual Model
+
+登船坡道替代信息化高墙。左侧高墙代表代码、系统、数据库门槛；右侧坡道代表业务理解、自然语言、任务拆解。坡道连接船票和 AI 航线。
+
+## Text Strategy
+
+- One title.
+- One central judgment.
+- Three short labels.
+- Avoid long explanatory sentences.
+
+## Continuity
+
+Use the accepted deep navy-black AI route operation-map base with cyan route glow and restrained gold highlights. Motif: 登船坡道 + 船票.
--- a/projects/2026-06-29-搭上AI这条船/slides/s10/s10-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s10/s10-v1-design.md
@ -0,0 +1,28 @@
+# s10-v1 Design
+
+## 页面
+
+P10：Agentic 技术平权：外行也能上船
+
+## 设计目标
+
+建立信心核心：这不是要求销售老兵变成程序员，而是让他们相信自己可以先进入 AI 探索轨。页面要把“技术门槛下降、业务理解上升”讲清楚。
+
+## 画面概念
+
+统一深蓝黑 AI 航线作战图底板。左侧是过去的信息化高墙，墙上有代码、系统、数据库的冷色符号；右侧是一条发光登船坡道，业务理解、自然语言、任务拆解三个节点沿坡道上船。坡道通向金色船票和远方 AI 航线。
+
+## 文案布局
+
+- 顶部标题：Agentic 技术平权：外行也能上船
+- 中心判断：不懂代码，不等于不能做 AI
+- 三个短标签：业务理解 / 自然语言 / 任务拆解
+- 轻量对比：过去高墙 / 现在坡道
+
+## 视觉要求
+
+要给销售信心，不要讲成技术判断课。避免代码崇拜、程序员英雄、机器人老师。画面重点是“坡道降低门槛”，不是“技术不重要”。
+
+## 后期提示
+
+如果文字不稳，优先保留标题、中心判断和三个坡道标签。
--- a/projects/2026-06-29-搭上AI这条船/slides/s10/s10-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s10/s10-v1-prompt.md
@ -0,0 +1,30 @@
+# s10-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P10 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted previous package radar/style sample, `slides/s09/s09-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual. Match the accepted samples: deep navy-black AI maritime route operation map, subtle ocean chart texture, compass/radar geometry, cyan signal lines, restrained gold highlights, large readable white Chinese title.
+
+Page title at top: “Agentic 技术平权：外行也能上船”.
+
+Main visual: left side shows an old high technical wall labeled “过去高墙”, with small muted symbols for “代码”, “系统”, “数据库”. Right side shows a bright cyan-gold boarding ramp replacing the wall, leading toward a glowing abstract AI route and a gold-edged boarding ticket. On the ramp place three clear labels: “业务理解”, “自然语言”, “任务拆解”.
+
+Center judgment in large readable text: “不懂代码，不等于不能做 AI”.
+
+Mood: confidence for experienced business people entering AI exploration. The page must communicate lowered entry barrier, not anti-technology. Avoid programmer worship, code walls as hero, robot teacher, customer proposal style, SaaS dashboard UI, government report, cartoon ship, real people or portraits. Use short, legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s11/s11-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s11/s11-v1-brief.md
@ -0,0 +1,32 @@
+# s11-v1 Brief
+
+## Page
+
+- Unit: s11
+- Source page: P11
+- Title: 传统软件工程师和 Agentic 工程师，是两个物种
+- Page family: Confidence / Agentic Entry
+- Version: v1
+
+## Core Conclusion
+
+这不是技术栈升级，而是能力系统变化。
+
+## Objective
+
+解释为什么客户 AI 价值探索不只是传统软件工程问题，并继续把销售老兵拉进探索前场。
+
+## Visual Model
+
+左右对比：确定性工程蓝图 vs 不确定性探索地图。中间是“能力系统变化”的分界线。
+
+## Text Strategy
+
+- One long title.
+- Two side labels.
+- One fairness sentence.
+- One bottom judgment.
+
+## Continuity
+
+Use the accepted deep navy-black AI route operation-map base with cyan and gold signal accents. Motif: capability-system contrast inside the Agentic entry family.
--- a/projects/2026-06-29-搭上AI这条船/slides/s11/s11-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s11/s11-v1-design.md
@ -0,0 +1,29 @@
+# s11-v1 Design
+
+## 页面
+
+P11：传统软件工程师和 Agentic 工程师，是两个物种
+
+## 设计目标
+
+强化反直觉判断：这不是技术栈升级，而是能力系统变化。页面必须服务“你们有机会进入前场”，不能变成贬低技术团队。
+
+## 画面概念
+
+统一深蓝黑 AI 航线作战图底板。左右对比：左侧是稳定工程管线、接口、数据库和蓝图；右侧是探索地图、任务网络、智能体链路。中间用一道发光分界线写“能力系统变化”。
+
+## 文案布局
+
+- 顶部标题：传统软件工程师和 Agentic 工程师，是两个物种
+- 左侧标题：传统软件工程师
+- 右侧标题：Agentic 工程师
+- 中心短句：不是谁高级谁低级，是两种不同能力系统
+- 底部判断：一个做稳确定需求，一个探索模糊问题
+
+## 视觉要求
+
+左右都要专业、克制、有价值感。左侧稳定，右侧探索。不要做成人物对立、职业嘲讽或工程架构课。
+
+## 后期提示
+
+如文字不稳，保留左右标题、中心短句和底部判断。
--- a/projects/2026-06-29-搭上AI这条船/slides/s11/s11-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s11/s11-v1-prompt.md
@ -0,0 +1,32 @@
+# s11-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P11 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted previous package radar/style sample, `slides/s09/s09-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual in the accepted style: deep navy-black AI maritime operation map, compass grid, cyan signal glow, restrained gold accents, crisp large Chinese typography.
+
+Page title at top: “传统软件工程师和 Agentic 工程师，是两个物种”.
+
+Layout: professional left-right contrast, not hostile. Left side label: “传统软件工程师”, visualized as stable engineering blueprint, pipes, API nodes, database blocks, test checkmarks, with short labels “确定需求”, “写代码”, “做系统”. Right side label: “Agentic 工程师”, visualized as exploration map, fuzzy target becoming task chain, AI agent nodes, validation loops, with short labels “理解目标”, “拆任务”, “验输出”.
+
+Center divider phrase in gold/cyan: “不是谁高级谁低级，是两种不同能力系统”.
+
+Bottom sentence: “一个做稳确定需求，一个探索模糊问题”.
+
+Keep both sides credible and useful. Avoid mocking programmers, dense architecture lecture, SaaS dashboard, customer proposal, government report, robots, real portraits, cartoon style. Use short readable Chinese only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s12/s12-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s12/s12-v1-brief.md
@ -0,0 +1,32 @@
+# s12-v1 Brief
+
+## Page
+
+- Unit: s12
+- Source page: P12
+- Title: 第一阶段要求：不是懂 AI，而是会带着 AI 工作
+- Page family: Confidence / Agentic Entry
+- Version: v1
+
+## Core Conclusion
+
+先会用，再会卖，再会带客户用。
+
+## Objective
+
+把“学 AI”的压力转成“带着 AI 工作”的第一阶段行动习惯。
+
+## Visual Model
+
+四节点 AI 工作闭环：会前准备 -> 现场观察 -> 会后复盘 -> 下一步行动。
+
+## Text Strategy
+
+- One title.
+- One central judgment.
+- Four loop labels.
+- One bottom action sentence.
+
+## Continuity
+
+Use the accepted deep navy-black AI route operation-map base with radar glow, route lines, cyan loop, and gold central judgment. Motif: 登船坡道 / AI work-loop route.
--- a/projects/2026-06-29-搭上AI这条船/slides/s12/s12-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s12/s12-v1-design.md
@ -0,0 +1,28 @@
+# s12-v1 Design
+
+## 页面
+
+P12：第一阶段要求：不是懂 AI，而是会带着 AI 工作
+
+## 设计目标
+
+降低进入门槛，让观众觉得今天晚上就能开始。核心是形成 AI 工作习惯，而不是先学半年原理。
+
+## 画面概念
+
+统一深蓝黑 AI 航线作战图底板。中心是一个发光闭环：会前准备、现场观察、会后复盘、下一步行动。闭环外层有雷达扫描和客户画像卡片，表示 AI 帮助销售在真实客户经营中形成循环。
+
+## 文案布局
+
+- 顶部标题：第一阶段要求：不是懂 AI，而是会带着 AI 工作
+- 中心金句：先会用，再会卖，再会带客户用
+- 四个闭环节点：会前准备 / 现场观察 / 会后复盘 / 下一步行动
+- 底部短句：把 AI 用到自己身上，才有资格带客户上船
+
+## 视觉要求
+
+要有行动感和可开始感，不要像课程大纲。节点必须清楚，文本少。
+
+## 后期提示
+
+如文字不稳，优先保留四个闭环节点和中心金句。
--- a/projects/2026-06-29-搭上AI这条船/slides/s12/s12-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s12/s12-v1-prompt.md
@ -0,0 +1,32 @@
+# s12-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P12 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted previous package radar/style sample, `slides/s09/s09-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual. Keep the same accepted deep navy-black AI route operation-map base, ocean chart lines, compass marks, cyan signal light, restrained gold emphasis, readable Chinese text.
+
+Page title at top: “第一阶段要求：不是懂 AI，而是会带着 AI 工作”.
+
+Main visual: a clean glowing circular work-loop diagram over a radar-map background. Four nodes around the loop: “会前准备”, “现场观察”, “会后复盘”, “下一步行动”. Add subtle customer-profile cards and radar scan marks around the loop, not as a SaaS dashboard but as maritime operation-map annotations.
+
+Center judgment: “先会用，再会卖，再会带客户用”.
+
+Bottom sentence: “把 AI 用到自己身上，才有资格带客户上船”.
+
+Mood: immediately actionable, low entry barrier, confident. Avoid course syllabus, training manual, dense text, robot teacher, generic software UI, customer proposal, government report, real people or portraits. Use short, legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s13/s13-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s13/s13-v1-brief.md
@ -0,0 +1,31 @@
+# s13-v1 Brief
+
+## Page
+
+- Unit: s13
+- Source page: P13
+- Title: 我们卖的不是 AI 产品，是一条企业智能化路径
+- Page family: Business Path / Customer Judgment
+- Version: v1
+
+## Core Conclusion
+
+客户真正需要的，是从想用 AI 到用好 AI 的路径。
+
+## Objective
+
+把团队要卖的东西从“课程 / 软件 / 系统”重新定义为企业智能化路径。
+
+## Visual Model
+
+航线节点地图：想用 AI -> 看懂 AI -> 找到场景 -> 形成方案 -> 试点验证 -> 系统落地 -> 用好 AI。
+
+## Text Strategy
+
+- One title.
+- Route node labels.
+- One bottom value sentence.
+
+## Continuity
+
+Use the accepted deep navy-black AI route operation-map base with cyan route and gold destination markers. Motif: AI 航线 / 港口 / 灯塔.
--- a/projects/2026-06-29-搭上AI这条船/slides/s13/s13-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s13/s13-v1-design.md
@ -0,0 +1,27 @@
+# s13-v1 Design
+
+## 页面
+
+P13：我们卖的不是 AI 产品，是一条企业智能化路径
+
+## 设计目标
+
+定义商业产品：客户真正需要的是从想用 AI 到用好 AI 的路径。页面要从信心段落转入业务路径段落。
+
+## 画面概念
+
+统一深蓝黑 AI 航线作战图底板。一条明亮航线从左侧“想用 AI”港口出发，经过看懂 AI、找到场景、形成方案、试点验证、系统落地，最终抵达右侧“用好 AI”港口。每个节点像灯塔或浮标。
+
+## 文案布局
+
+- 顶部标题：我们卖的不是 AI 产品，是一条企业智能化路径
+- 航线节点：想用 AI / 看懂 AI / 找到场景 / 形成方案 / 试点验证 / 系统落地 / 用好 AI
+- 底部短句：产品是入口，路径才是价值
+
+## 视觉要求
+
+路径要一眼成立，不要做成产品清单或客户方案页。整体保持内训启动会的作战地图感。
+
+## 后期提示
+
+如节点文字过多，优先保留“想用 AI -> 用好 AI”和底部短句。
--- a/projects/2026-06-29-搭上AI这条船/slides/s13/s13-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s13/s13-v1-prompt.md
@ -0,0 +1,30 @@
+# s13-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P13 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted previous package radar/style sample, `slides/s09/s09-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual in the same accepted style: deep navy-black AI maritime route operation map, ocean chart texture, subtle compass geometry, cyan glowing route, restrained gold action nodes, large readable Chinese typography.
+
+Page title at top: “我们卖的不是 AI 产品，是一条企业智能化路径”.
+
+Main visual: a clear maritime route map from left to right. Start port label: “想用 AI”. Route nodes as glowing buoys or small lighthouses: “看懂 AI”, “找到场景”, “形成方案”, “试点验证”, “系统落地”. Final destination port label: “用好 AI”. The route should feel practical and guided, not like a software product roadmap.
+
+Bottom sentence: “产品是入口，路径才是价值”.
+
+Mood: practical navigation model for clients. Avoid product catalog, sales proposal, SaaS dashboard, government report, dense process chart, cartoon ship, robots, real people. Use short, legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s14/s14-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s14/s14-v1-brief.md
@ -0,0 +1,32 @@
+# s14-v1 Brief
+
+## Page
+
+- Unit: s14
+- Source page: P14
+- Title: 三级服务路径：培训 → 顾问 → 系统
+- Page family: Business Path / Customer Judgment
+- Version: v1
+
+## Core Conclusion
+
+培训是入口，顾问是筛选，系统是重服务。
+
+## Objective
+
+把三级服务路径可视化为团队可复述的业务基础模型。
+
+## Visual Model
+
+三段航线：近海训练区 -> 航道判断区 -> 深水建设区，对应培训、顾问、系统。
+
+## Text Strategy
+
+- One title.
+- Three large stage labels.
+- One short label per stage.
+- One bottom warning sentence.
+
+## Continuity
+
+Use the accepted deep navy-black AI route operation-map base. Motif: AI 航线 + service-stage nodes.
--- a/projects/2026-06-29-搭上AI这条船/slides/s14/s14-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s14/s14-v1-design.md
@ -0,0 +1,30 @@
+# s14-v1 Design
+
+## 页面
+
+P14：三级服务路径：培训 → 顾问 → 系统
+
+## 设计目标
+
+展示核心商业路径，让这页成为团队未来出去讲业务的基础模型。核心是培训入口、顾问筛选、系统重服务。
+
+## 画面概念
+
+统一深蓝黑 AI 航线作战图底板。三段式航线从近海训练区进入航道判断区，再到深水建设区。每段有一个醒目节点：培训、顾问、系统，并配短标签说明角色。
+
+## 文案布局
+
+- 顶部标题：三级服务路径：培训 → 顾问 → 系统
+- 三段节点：
+  - 培训：建立认知
+  - 顾问：判断路径
+  - 系统：重服务交付
+- 底部金句：不是所有客户都直接上系统
+
+## 视觉要求
+
+要像核心业务模型，不要像课程表。系统段要显得更重、更严，但不要压抑。
+
+## 后期提示
+
+如文字不稳，保留三个大节点和底部金句。
--- a/projects/2026-06-29-搭上AI这条船/slides/s14/s14-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s14/s14-v1-prompt.md
@ -0,0 +1,30 @@
+# s14-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P14 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted previous package radar/style sample, `slides/s09/s09-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual, matching the accepted deep navy-black AI maritime operation-map style, with cyan route glow, gold stage markers, compass lines, and readable Chinese typography.
+
+Page title at top: “三级服务路径：培训 → 顾问 → 系统”.
+
+Main visual: a three-stage maritime route from left to right or rising diagonally. Stage 1 nearshore training zone: large label “培训”, sublabel “建立认知”. Stage 2 channel judgment zone: large label “顾问”, sublabel “判断路径”. Stage 3 deep-water construction zone: large label “系统”, sublabel “重服务交付”. Use route nodes, depth zones, and subtle lock/channel geometry, not a generic staircase clipart.
+
+Bottom sentence: “不是所有客户都直接上系统”.
+
+Mood: core business model, practical and memorable. Avoid course timetable, customer proposal, government report, SaaS dashboard, dense table, cartoon ship, robots, real people. Use short, legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s15/s15-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s15/s15-v1-brief.md
@ -0,0 +1,32 @@
+# s15-v1 Brief
+
+## Page
+
+- Unit: s15
+- Source page: P15
+- Title: 培训可以宽，顾问要筛，系统必须严
+- Page family: Business Path / Customer Judgment
+- Version: v1
+
+## Core Conclusion
+
+做错项目，比少做项目更伤。
+
+## Objective
+
+让销售明白客户推进要分层筛选，尤其不能把不成熟客户直接推系统。
+
+## Visual Model
+
+航道闸门 / 漏斗：培训宽入口，顾问筛选，系统严闸门。
+
+## Text Strategy
+
+- One title.
+- Three strict short labels.
+- One central judgment.
+- One bottom discipline sentence.
+
+## Continuity
+
+Use the accepted deep navy-black AI route operation-map base. Motif: 航道闸门 / funnel gate.
--- a/projects/2026-06-29-搭上AI这条船/slides/s15/s15-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s15/s15-v1-design.md
@ -0,0 +1,28 @@
+# s15-v1 Design
+
+## 页面
+
+P15：培训可以宽，顾问要筛，系统必须严
+
+## 设计目标
+
+建立客户推进纪律：宽入口、严系统。页面要稳，但不能把兴奋感拉成纪律教育。
+
+## 画面概念
+
+统一深蓝黑 AI 航线作战图底板。画面中心是三段航道闸门 / 漏斗：培训入口较宽，顾问通道收窄，系统闸门最窄、最亮、最严格。旁边有风险提示，不做错项目。
+
+## 文案布局
+
+- 顶部标题：培训可以宽，顾问要筛，系统必须严
+- 三段标签：培训：宽入口 / 顾问：要筛选 / 系统：必须严
+- 中心判断：做错项目，比少做项目更伤
+- 底部金句：宁可少做项目，也不能做错项目
+
+## 视觉要求
+
+闸门要有纪律感，但不要恐吓。要体现筛选客户也是销售价值。
+
+## 后期提示
+
+如文字不稳，保留三段标签和底部金句。
--- a/projects/2026-06-29-搭上AI这条船/slides/s15/s15-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s15/s15-v1-prompt.md
@ -0,0 +1,30 @@
+# s15-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P15 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted previous package radar/style sample, `slides/s09/s09-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual in the accepted deep navy-black AI maritime route-map style, with cyan signal lines, restrained gold warning/action accents, compass grid, and bold readable Chinese typography.
+
+Page title at top: “培训可以宽，顾问要筛，系统必须严”.
+
+Main visual: a maritime lock-gate funnel. Top wide gate labeled “培训：宽入口”. Middle narrower gate labeled “顾问：要筛选”. Bottom strict narrow gate labeled “系统：必须严”. The gate geometry should feel like a channel checkpoint on an AI route map, not a generic sales funnel. Add a concise central judgment plaque: “做错项目，比少做项目更伤”.
+
+Bottom sentence: “宁可少做项目，也不能做错项目”.
+
+Mood: disciplined but not scary, practical customer filtering. Avoid punishment board, compliance training, government report, customer proposal, spreadsheet funnel, SaaS dashboard, cartoon ship, robots, real people. Use short, legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s16/s16-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s16/s16-v1-brief.md
@ -0,0 +1,32 @@
+# s16-v1 Brief
+
+## Page
+
+- Unit: s16
+- Source page: P16
+- Title: 客户适配判断模型
+- Page family: Business Path / Customer Judgment
+- Version: v1
+
+## Core Conclusion
+
+客户能不能做 AI，至少看四件事。
+
+## Objective
+
+给销售一个初步判断客户是否适合智能化的四方向罗盘。
+
+## Visual Model
+
+四象限罗盘 + 雷达：基础建设、组织认知、场景价值、权力结构。
+
+## Text Strategy
+
+- One title.
+- One central judgment.
+- Four short direction labels.
+- One bottom caution sentence.
+
+## Continuity
+
+Use the accepted deep navy-black AI route operation-map base. Motif: 罗盘 + 雷达.
--- a/projects/2026-06-29-搭上AI这条船/slides/s16/s16-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s16/s16-v1-design.md
@ -0,0 +1,28 @@
+# s16-v1 Design
+
+## 页面
+
+P16：客户适配判断模型
+
+## 设计目标
+
+给销售一个初步判断客户的罗盘。页面信息可能多，必须把关键词压缩成四个方向。
+
+## 画面概念
+
+统一深蓝黑 AI 航线作战图底板。中心是“客户是否适合智能化”的罗盘 / 雷达盘，四个方向分别是基础建设、组织认知、场景价值、权力结构。每个方向只放少量关键词。
+
+## 文案布局
+
+- 顶部标题：客户适配判断模型
+- 中心判断：客户能不能做 AI，至少看四件事
+- 四向罗盘：基础建设 / 组织认知 / 场景价值 / 权力结构
+- 底部短句：客户想不想做，不等于客户能不能做
+
+## 视觉要求
+
+像销售未来可用的判断工具，不要像表格堆砌。罗盘和雷达结合，四个方向清楚。
+
+## 后期提示
+
+如文字不稳，优先保留四个方向和底部短句。
--- a/projects/2026-06-29-搭上AI这条船/slides/s16/s16-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s16/s16-v1-prompt.md
@ -0,0 +1,32 @@
+# s16-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: required project load order plus P16 skeleton, visual system, style master, motif registry, task plan, production status.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted previous package radar/style sample, `slides/s09/s09-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual in the accepted style: deep navy-black AI maritime operation map, radar scan, compass rose, subtle ocean chart, cyan signal glow, restrained gold highlights, crisp readable Chinese typography.
+
+Page title at top: “客户适配判断模型”.
+
+Main visual: a large four-direction compass combined with radar. Center label: “客户是否适合智能化”. Four compass directions with clear labels: “基础建设”, “组织认知”, “场景价值”, “权力结构”. Add tiny supporting keywords around each direction, very short only: “系统/数据”, “老板/员工”, “高频/高成本”, “决策/付款”. Keep the compass clean and usable, not a crowded table.
+
+Central or upper judgment: “客户能不能做 AI，至少看四件事”.
+
+Bottom sentence: “客户想不想做，不等于客户能不能做”.
+
+Mood: practical judgment tool for experienced salespeople. Avoid dense table, government report, customer proposal, SaaS dashboard, technical architecture lecture, cartoon ship, robots, real portraits. Use short, legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s17/s17-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s17/s17-v1-brief.md
@ -0,0 +1,33 @@
+# s17-v1 Brief
+
+## Execution Card
+
+- Unit: s17
+- Source page: P17
+- Title: 双轨体系：先探索价值，再工程兑现
+- Family: Business Path / Customer Judgment
+- Motif: 双轨 / 双航道
+- Running mode: prompt and generate
+- Review mode: none
+- Version: s17-v1
+- Overwrite allowed: no
+
+## Source Judgment
+
+眼下重点不是系统建设分工，而是先把探索轨跑起来。
+
+## Visual Direction
+
+Use the accepted deep navy-black AI maritime route operation-map base. Show two coordinated channels: a bright exploration track above and a steadier engineering track below. The exploration vessel goes first to identify value; the engineering vessel follows only after value is confirmed.
+
+## Text To Keep Short
+
+- 双轨体系：先探索价值，再工程兑现
+- 探索轨：发现价值
+- 工程轨：兑现价值
+- 先拉上船，再说怎么造大船
+- 我们现在要先学会发现价值
+
+## Avoid
+
+Do not drift into engineering management, system architecture, SaaS dashboard, customer proposal, cartoon ship, humanoid robots, or concrete people.
--- a/projects/2026-06-29-搭上AI这条船/slides/s17/s17-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s17/s17-v1-design.md
@ -0,0 +1,27 @@
+# s17-v1 Design
+
+## 页面定位
+
+- 页面：P17 / s17
+- 标题：双轨体系：先探索价值，再工程兑现
+- 页面家族：Business Path / Customer Judgment
+- 核心结论：眼下重点不是系统建设分工，而是先把探索轨跑起来。
+- 核心母题：双轨 / 双航道
+
+## 视觉概念
+
+在统一的深蓝黑 AI 航线作战图底盘上，画面横向展开两条并行航道。上方“探索轨”用更亮的青色航线表现，像一艘雷达探测船先行探水深；下方“工程轨”用更深、更稳的蓝灰航道表现，像工程船队在后方等待兑现价值。
+
+两条轨道不是并列抢戏，而是有先后关系：探索轨前置、工程轨后接。中间用金色转接节点强调“价值确认后再工程兑现”。
+
+## 文字与布局
+
+- 顶部主标题：双轨体系：先探索价值，再工程兑现
+- 中央大判断：先拉上船，再说怎么造大船
+- 上轨标签：探索轨 / 发现价值
+- 下轨标签：工程轨 / 兑现价值
+- 底部短句：我们现在要先学会发现价值
+
+## 风格约束
+
+沿用 s02 / s16 的深蓝黑航海作战图、雷达、罗盘线、青色信号和克制金色强调。不要做成工程管理架构图、铁路交通图、SaaS 流程图或客户方案页。
--- a/projects/2026-06-29-搭上AI这条船/slides/s17/s17-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s17/s17-v1-prompt.md
@ -0,0 +1,33 @@
+# s17-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false,"base_url":"https://api.openai.com/v1","model":"gpt-image-2","env_flag_value":"(unset)"}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\poster-and-campaigns\campaign-kv.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: AGENTS.md, formal production handoff, project.md, execution-plan.md, slides.md, visual-system.md, style-prompt-master.md, motif-registry.md, P1-P24 skeleton, task-plan.md, production-run-status.md.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted customer-judgment style sample, `slides/s16/s16-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual matching Image 1 and Image 2: deep navy-black AI maritime operation map, subtle ocean chart grid, radar glow, compass lines, cyan signal routes, restrained gold decision highlights, crisp readable Chinese typography.
+
+Page title at top: “双轨体系：先探索价值，再工程兑现”.
+
+Main visual: two parallel maritime channels across the slide. Upper channel is bright cyan and labeled “探索轨：发现价值”, with a radar scout vessel moving first through customer interview, training, scenario sorting, agent prototype, pilot validation nodes shown as tiny route lights. Lower channel is darker, steadier blue-gray and labeled “工程轨：兑现价值”, with an engineering vessel following later through system, database, permission, interface, deployment, testing nodes as subdued markers. Add one gold transfer node between tracks labeled “价值确认”.
+
+Central large judgment plaque: “先拉上船，再说怎么造大船”.
+
+Bottom sentence: “我们现在要先学会发现价值”.
+
+Mood: practical navigation model, sales veterans enter the exploration track first. Avoid engineering management chart, system architecture lecture, SaaS dashboard, customer proposal, government report, cyberpunk poster, cartoon ship, robots, stock people, concrete portraits. Use short legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s18/s18-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s18/s18-v1-brief.md
@ -0,0 +1,35 @@
+# s18-v1 Brief
+
+## Execution Card
+
+- Unit: s18
+- Source page: P18
+- Title: 新团队协作结构
+- Family: Team Structure / Boundaries
+- Motif: 船舱 / 作战地图
+- Running mode: prompt and generate
+- Review mode: none
+- Version: s18-v1
+- Overwrite allowed: no
+
+## Source Judgment
+
+未来不是老板 + 销售 + 技术，而是四类角色协同。
+
+## Visual Direction
+
+Use a ship-compartment operation map. Put “客户智能化旅程” in the center and four functional compartments around it: chief expert, customer operations, AI business engineer, deterministic engineering foundation.
+
+## Text To Keep Short
+
+- 新团队协作结构
+- 客户智能化旅程
+- 首席专家
+- 客户经营
+- AI业务工程
+- 工程底座
+- 各就各位，才能打大仗
+
+## Avoid
+
+No HR org chart, no portrait wall, no policy page, no customer proposal, no generic SaaS dashboard.
--- a/projects/2026-06-29-搭上AI这条船/slides/s18/s18-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s18/s18-v1-design.md
@ -0,0 +1,26 @@
+# s18-v1 Design
+
+## 页面定位
+
+- 页面：P18 / s18
+- 标题：新团队协作结构
+- 页面家族：Team Structure / Boundaries
+- 核心结论：未来不是老板 + 销售 + 技术，而是四类角色协同。
+- 核心母题：船舱 / 作战地图
+
+## 视觉概念
+
+统一底盘上呈现一张“船舱式作战地图”。中心不是老板，而是“客户智能化旅程”；四个功能舱位环绕中心：学者型首席专家、客户经营负责人、AI 业务工程师、确定性工程底座。
+
+每个舱位使用抽象图标和短标签，连接线汇入中心航线，表达“各就各位，才能打大仗”。画面要有组织感但不僵硬，像启动会的作战分工图，不像 HR 组织架构。
+
+## 文字与布局
+
+- 顶部主标题：新团队协作结构
+- 中心：客户智能化旅程
+- 四舱位：首席专家 / 客户经营 / AI业务工程 / 工程底座
+- 底部短句：各就各位，才能打大仗
+
+## 风格约束
+
+保持深蓝黑、青色航线、金色角色节点。避免正式组织架构图、职级汇报页、团队头像墙和政府报告风。
--- a/projects/2026-06-29-搭上AI这条船/slides/s18/s18-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s18/s18-v1-prompt.md
@ -0,0 +1,35 @@
+# s18-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false,"base_url":"https://api.openai.com/v1","model":"gpt-image-2","env_flag_value":"(unset)"}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\poster-and-campaigns\campaign-kv.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: AGENTS.md, formal production handoff, project.md, execution-plan.md, slides.md, visual-system.md, style-prompt-master.md, motif-registry.md, P1-P24 skeleton, task-plan.md, production-run-status.md.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted customer-judgment style sample, `slides/s16/s16-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual matching Image 1 and Image 2: deep navy-black AI maritime operation map, cyan route glow, radar and compass geometry, restrained gold role highlights, crisp readable Chinese typography.
+
+Page title at top: “新团队协作结构”.
+
+Main visual: a ship-compartment operation map, not a stiff org chart. Center glowing route node labeled “客户智能化旅程”. Four surrounding compartments connected to the center:
+1. “首席专家” sublabel “方法论 / 判断”
+2. “客户经营” sublabel “关系 / 初筛”
+3. “AI业务工程” sublabel “转译 / 原型”
+4. “工程底座” sublabel “系统 / 交付”
+
+Use compartment panels, route lines, subtle deck blueprint geometry, and gold-cyan node links. Add bottom sentence: “各就各位，才能打大仗”.
+
+Mood: brotherly trust plus company discipline, practical collaboration map. Avoid HR org chart, portrait wall, meeting-room stock photo, customer proposal, government report, SaaS dashboard, cyberpunk poster, robots, cartoon ship. Use short legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s19/s19-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s19/s19-v1-brief.md
@ -0,0 +1,32 @@
+# s19-v1 Brief
+
+## Execution Card
+
+- Unit: s19
+- Source page: P19
+- Title: 我们是帮客户上船的人
+- Family: Team Structure / Boundaries
+- Motif: 角色图标 / 登船路径
+- Running mode: prompt and generate
+- Review mode: none
+- Version: s19-v1
+- Overwrite allowed: no
+
+## Source Judgment
+
+客户不知道怎么上 AI 这条船，我们就是带路的人。
+
+## Visual Direction
+
+Create an easy-to-remember role map around an abstract boarding path. Five icons: entrance, radar, translator, propeller, firewall. Keep it simple and iconic.
+
+## Text To Keep Short
+
+- 我们是帮客户上船的人
+- 客户不知道怎么上 AI 这条船，我们就是那个带路的人
+- 入口 / 雷达 / 翻译器 / 推进器 / 防火墙
+- 找到客户，听懂客户，判断客户，推动客户，守住边界
+
+## Avoid
+
+No real people, no cartoon mascot, no customer brochure, no motivational poster, no humanoid robots.
--- a/projects/2026-06-29-搭上AI这条船/slides/s19/s19-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s19/s19-v1-design.md
@ -0,0 +1,26 @@
+# s19-v1 Design
+
+## 页面定位
+
+- 页面：P19 / s19
+- 标题：我们是帮客户上船的人
+- 页面家族：Team Structure / Boundaries
+- 核心结论：客户不知道怎么上 AI 这条船，我们就是带路的人。
+- 核心母题：五个角色图标围绕登船路径
+
+## 视觉概念
+
+这页要成为容易记住的精神锚点。画面中心是一条抽象登船路径，从客户侧通向 AI 航线。五个角色图标围绕路径排列：入口、雷达、翻译器、推进器、防火墙。
+
+视觉语言要简洁、有力量，不做复杂插画。每个角色用一个金色或青色图标表示，整体像一张“客户上船行动图”。
+
+## 文字与布局
+
+- 顶部主标题：我们是帮客户上船的人
+- 中央金句：客户不知道怎么上 AI 这条船，我们就是那个带路的人
+- 五角色：入口 / 雷达 / 翻译器 / 推进器 / 防火墙
+- 底部短句：找到客户，听懂客户，判断客户，推动客户，守住边界
+
+## 风格约束
+
+避免人物肖像、卡通船、吉祥物和励志海报。必须像内部作战图中的角色地图。
--- a/projects/2026-06-29-搭上AI这条船/slides/s19/s19-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s19/s19-v1-prompt.md
@ -0,0 +1,38 @@
+# s19-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false,"base_url":"https://api.openai.com/v1","model":"gpt-image-2","env_flag_value":"(unset)"}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\poster-and-campaigns\campaign-kv.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: AGENTS.md, formal production handoff, project.md, execution-plan.md, slides.md, visual-system.md, style-prompt-master.md, motif-registry.md, P1-P24 skeleton, task-plan.md, production-run-status.md.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted customer-judgment style sample, `slides/s16/s16-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual matching Image 1 and Image 2: deep navy-black AI maritime operation map, cyan route glow, radar scan, compass grid, restrained gold highlights, crisp readable Chinese typography.
+
+Page title at top: “我们是帮客户上船的人”.
+
+Main visual: an abstract boarding path across a dark AI route map, leading from “客户” toward a glowing AI route. Around the path place five clean role icons with short labels:
+1. “入口” with a boarding-ticket icon
+2. “雷达” with a radar-scan icon
+3. “翻译器” with two-way arrows
+4. “推进器” with a propulsion engine symbol
+5. “防火墙” with a shield gate
+
+Central large quote: “客户不知道怎么上 AI 这条船，我们就是那个带路的人”.
+
+Bottom sentence: “找到客户，听懂客户，判断客户，推动客户，守住边界”.
+
+Mood: memorable spiritual anchor, simple role map, internal route-operation energy. Avoid real people, portraits, cute cartoon ship, mascot, customer brochure, government report, SaaS dashboard, cyberpunk poster, robots, dense text. Use short legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s20/s20-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s20/s20-v1-brief.md
@ -0,0 +1,34 @@
+# s20-v1 Brief
+
+## Execution Card
+
+- Unit: s20
+- Source page: P20
+- Title: 魏征机制：你们也是我的现实校准器
+- Family: Team Structure / Boundaries
+- Motif: 校准罗盘
+- Running mode: prompt and generate
+- Review mode: none
+- Version: s20-v1
+- Overwrite allowed: no
+
+## Source Judgment
+
+你们不是打断我，是帮我保护会议节奏。
+
+## Visual Direction
+
+Use a compass correction metaphor. A route drifts off course, then three calibration actions return it to the main route: before meeting reminder, during meeting pullback, after meeting debrief.
+
+## Text To Keep Short
+
+- 魏征机制：你们也是我的现实校准器
+- 会前提醒
+- 会中拉回
+- 会后复盘
+- 你们不是打断我，是帮我保护会议节奏
+- AI 能提醒我，但现实里的魏征更管用
+
+## Avoid
+
+No historical portrait, no ancient costume style, no rules board, no HR process, no government report.
--- a/projects/2026-06-29-搭上AI这条船/slides/s20/s20-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s20/s20-v1-design.md
@ -0,0 +1,26 @@
+# s20-v1 Design
+
+## 页面定位
+
+- 页面：P20 / s20
+- 标题：魏征机制：你们也是我的现实校准器
+- 页面家族：Team Structure / Boundaries
+- 核心结论：你们不是打断我，是帮我保护会议节奏。
+- 核心母题：校准罗盘 / 航向修正
+
+## 视觉概念
+
+画面以一个正在被校准的罗盘为中心。偏航的航线从左侧进入，经过“会前提醒 / 会中拉回 / 会后复盘”三个校准动作后，重新回到主航线。
+
+这页要有人情味但不能出现具体人物。用“现实校准器”这个隐喻来表达兄弟间的提醒和协作，不做成管理制度页。
+
+## 文字与布局
+
+- 顶部主标题：魏征机制：你们也是我的现实校准器
+- 中央判断：你们不是打断我，是帮我保护会议节奏
+- 三段：会前提醒 / 会中拉回 / 会后复盘
+- 底部金句：AI 能提醒我，但现实里的魏征更管用
+
+## 风格约束
+
+保持航线图与罗盘感。避免历史人物画像、古风海报、制度板、会议管理流程图。
--- a/projects/2026-06-29-搭上AI这条船/slides/s20/s20-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s20/s20-v1-prompt.md
@ -0,0 +1,33 @@
+# s20-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false,"base_url":"https://api.openai.com/v1","model":"gpt-image-2","env_flag_value":"(unset)"}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\poster-and-campaigns\campaign-kv.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: AGENTS.md, formal production handoff, project.md, execution-plan.md, slides.md, visual-system.md, style-prompt-master.md, motif-registry.md, P1-P24 skeleton, task-plan.md, production-run-status.md.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted customer-judgment style sample, `slides/s16/s16-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual matching Image 1 and Image 2: deep navy-black AI maritime operation map, compass geometry, radar glow, cyan route lines, restrained gold correction markers, crisp readable Chinese typography.
+
+Page title at top: “魏征机制：你们也是我的现实校准器”.
+
+Main visual: a compass calibration and route-correction map. A cyan route line begins slightly off course, passes through three gold-cyan calibration nodes, then returns to the main route. The three nodes are labeled “会前提醒”, “会中拉回”, “会后复盘”. Center compass label: “现实校准器”.
+
+Large central judgment: “你们不是打断我，是帮我保护会议节奏”.
+
+Bottom sentence: “AI 能提醒我，但现实里的魏征更管用”.
+
+Mood: trusted internal feedback mechanism, brotherly but disciplined. Avoid historical portraits, ancient Chinese costume style, rules board, HR process page, government report, customer proposal, SaaS dashboard, cyberpunk poster, robots, real people. Use short legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s21/s21-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s21/s21-v1-brief.md
@ -0,0 +1,35 @@
+# s21-v1 Brief
+
+## Execution Card
+
+- Unit: s21
+- Source page: P21
+- Title: L0-L4 成长路径
+- Family: Team Structure / Boundaries
+- Motif: L0-L4 阶梯 / 航行资格
+- Running mode: prompt and generate
+- Review mode: none
+- Version: s21-v1
+- Overwrite allowed: no
+
+## Source Judgment
+
+头衔不是先给的，是训练和行为拿到的。
+
+## Visual Direction
+
+Use an upward navigation-license ladder, from L0 to L4. Each step is a clear growth station, not a ranking wall.
+
+## Text To Keep Short
+
+- L0-L4 成长路径
+- L0 渠道伙伴
+- L1 AI商务助理
+- L2 客户经营负责人
+- L3 商务负责人
+- L4 AI前端部署顾问
+- 不是所有人一步到位，但每个人都有上升通道
+
+## Avoid
+
+No school grading, no pressure ranking, no HR promotion chart, no stiff policy page.
--- a/projects/2026-06-29-搭上AI这条船/slides/s21/s21-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s21/s21-v1-design.md
@ -0,0 +1,26 @@
+# s21-v1 Design
+
+## 页面定位
+
+- 页面：P21 / s21
+- 标题：L0-L4 成长路径
+- 页面家族：Team Structure / Boundaries
+- 核心结论：头衔不是先给的，是训练和行为拿到的。
+- 核心母题：L0-L4 阶梯 / 航行资格
+
+## 视觉概念
+
+画面用左低右高的航行资格阶梯。每一级像一张航海许可或训练等级牌，而不是等级压迫。L0 到 L4 沿着青色路线逐步点亮，最终通向前端部署顾问。
+
+整体要给希望，不给压力。L0 不显得低人一等，而是“起点清楚”。
+
+## 文字与布局
+
+- 顶部主标题：L0-L4 成长路径
+- 中央判断：头衔不是先给的，是训练和行为拿到的
+- 五级：L0 渠道伙伴 / L1 AI商务助理 / L2 客户经营负责人 / L3 商务负责人 / L4 AI前端部署顾问
+- 底部短句：不是所有人一步到位，但每个人都有上升通道
+
+## 风格约束
+
+保持深蓝黑航线和金色资格节点。避免学校成绩单、KPI 压迫榜、HR 晋升制度页。
--- a/projects/2026-06-29-搭上AI这条船/slides/s21/s21-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s21/s21-v1-prompt.md
@ -0,0 +1,38 @@
+# s21-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false,"base_url":"https://api.openai.com/v1","model":"gpt-image-2","env_flag_value":"(unset)"}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\poster-and-campaigns\campaign-kv.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: AGENTS.md, formal production handoff, project.md, execution-plan.md, slides.md, visual-system.md, style-prompt-master.md, motif-registry.md, P1-P24 skeleton, task-plan.md, production-run-status.md.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted customer-judgment style sample, `slides/s16/s16-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual matching Image 1 and Image 2: deep navy-black AI maritime operation map, subtle compass grid, cyan route glow, restrained gold milestone nodes, crisp readable Chinese typography.
+
+Page title at top: “L0-L4 成长路径”.
+
+Main visual: an upward navigation-license ladder from left-low to right-high across the dark route map. Five illuminated stations, each like a clean maritime qualification plate:
+L0 “渠道伙伴”
+L1 “AI商务助理”
+L2 “客户经营负责人”
+L3 “商务负责人”
+L4 “AI前端部署顾问”
+
+Add a large central judgment: “头衔不是先给的，是训练和行为拿到的”.
+
+Bottom sentence: “不是所有人一步到位，但每个人都有上升通道”.
+
+Mood: hopeful growth path, disciplined but not pressure-heavy. Avoid school grading chart, HR promotion chart, pressure ranking, compliance policy page, customer proposal, government report, SaaS dashboard, cartoon ship, robots, real people. Use short legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s22/s22-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s22/s22-v1-brief.md
@ -0,0 +1,33 @@
+# s22-v1 Brief
+
+## Execution Card
+
+- Unit: s22
+- Source page: P22
+- Title: 今天先记住三条
+- Family: Team Structure / Boundaries
+- Motif: 三道安全闸
+- Running mode: prompt and generate
+- Review mode: none
+- Version: s22-v1
+- Overwrite allowed: no
+
+## Source Judgment
+
+规则可以慢慢细化，但底线今天先钉住。
+
+## Visual Direction
+
+Use three safety gates / seals in the accepted maritime operation-map style. The page should feel firm and lightweight, not punitive.
+
+## Text To Keep Short
+
+- 今天先记住三条
+- 不装专家
+- 不乱承诺
+- 不直接推系统
+- 老江湖也要适应新边界
+
+## Avoid
+
+No punishment board, no red compliance warning, no dense rule list, no government report.
--- a/projects/2026-06-29-搭上AI这条船/slides/s22/s22-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s22/s22-v1-design.md
@ -0,0 +1,26 @@
+# s22-v1 Design
+
+## 页面定位
+
+- 页面：P22 / s22
+- 标题：今天先记住三条
+- 页面家族：Team Structure / Boundaries
+- 核心结论：规则可以慢慢细化，但底线今天先钉住。
+- 核心母题：三道安全闸
+
+## 视觉概念
+
+画面呈现三道轻量但明确的航行安全闸门。每道闸门像一枚发光印章或检查点，不像纪律处分板。三条底线并列：不装专家、不乱承诺、不把不成熟客户直接推系统。
+
+整体氛围要稳，但不能把启动会兴奋感拉成规则教育。
+
+## 文字与布局
+
+- 顶部主标题：今天先记住三条
+- 中央判断：规则可以慢慢细化，但底线今天先钉住
+- 三卡：不装专家 / 不乱承诺 / 不直接推系统
+- 底部短句：老江湖也要适应新边界
+
+## 风格约束
+
+用航道闸门和金色底线印章；避免处罚、红色警示、制度墙、密密麻麻清单。
--- a/projects/2026-06-29-搭上AI这条船/slides/s22/s22-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s22/s22-v1-prompt.md
@ -0,0 +1,36 @@
+# s22-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false,"base_url":"https://api.openai.com/v1","model":"gpt-image-2","env_flag_value":"(unset)"}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\poster-and-campaigns\campaign-kv.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: AGENTS.md, formal production handoff, project.md, execution-plan.md, slides.md, visual-system.md, style-prompt-master.md, motif-registry.md, P1-P24 skeleton, task-plan.md, production-run-status.md.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted customer-judgment style sample, `slides/s16/s16-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual matching Image 1 and Image 2: deep navy-black AI maritime operation map, cyan route lines, compass grid, restrained gold safety-gate highlights, crisp readable Chinese typography.
+
+Page title at top: “今天先记住三条”.
+
+Main visual: three large clean safety gates or gold-cyan seals aligned across an AI route channel. Each gate is firm but not punitive:
+1. “不装专家” sublabel “不懂可以学”
+2. “不乱承诺” sublabel “不能承诺结果”
+3. “不直接推系统” sublabel “成熟客户再进”
+
+Central judgment: “规则可以慢慢细化，但底线今天先钉住”.
+
+Bottom sentence: “老江湖也要适应新边界”.
+
+Mood: lightweight boundary-setting, steady and practical, still part of an exciting kickoff. Avoid punishment board, red compliance warning, dense policy wall, government report, customer proposal, SaaS dashboard, cyberpunk poster, cartoon ship, robots, real people. Use short legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s23/s23-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s23/s23-v1-brief.md
@ -0,0 +1,34 @@
+# s23-v1 Brief
+
+## Execution Card
+
+- Unit: s23
+- Source page: P23
+- Title: 一个月训练：成为 AI 增强型客户经营人员
+- Family: Training / Action
+- Motif: 训练路线图
+- Running mode: prompt and generate
+- Review mode: none
+- Version: s23-v1
+- Overwrite allowed: no
+
+## Source Judgment
+
+一个月后，大家至少要会用 AI 做准备、复盘和客户判断。
+
+## Visual Direction
+
+Use a four-week route map across the accepted dark AI maritime operation base. Keep each week as one clear node with a short training target.
+
+## Text To Keep Short
+
+- 一个月训练：成为 AI 增强型客户经营人员
+- 第一周：和 AI 交朋友
+- 第二周：客户沟通与业务理解
+- 第三周：产品边界与承诺规则
+- 第四周：模拟拜访与认证
+- 不是培养技术专家，是培养 AI 增强型商务人员
+
+## Avoid
+
+No course syllabus, no school homework look, no formal certificate poster, no dense schedule table.
--- a/projects/2026-06-29-搭上AI这条船/slides/s23/s23-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s23/s23-v1-design.md
@ -0,0 +1,26 @@
+# s23-v1 Design
+
+## 页面定位
+
+- 页面：P23 / s23
+- 标题：一个月训练：成为 AI 增强型客户经营人员
+- 页面家族：Training / Action
+- 核心结论：一个月后，大家至少要会用 AI 做准备、复盘和客户判断。
+- 核心母题：训练路线图
+
+## 视觉概念
+
+画面是一条四周训练航线，从“新船员训练”出发，到“AI 增强型客户经营人员”。四个周节点沿路线排布，节点文字只保留目标，不展开课程大纲。
+
+这页从结构页转入行动页，要更明确、更收束，但仍沿用统一航线作战图底盘。
+
+## 文字与布局
+
+- 顶部主标题：一个月训练：成为 AI 增强型客户经营人员
+- 中央判断：一个月后，大家至少要会用 AI 做准备、复盘和客户判断
+- 四周：和 AI 交朋友 / 客户沟通与业务理解 / 产品边界与承诺规则 / 模拟拜访与认证
+- 底部金句：不是培养技术专家，是培养 AI 增强型商务人员
+
+## 风格约束
+
+避免课程表、学校培训海报、密集 syllabus。必须是行动路线图。
--- a/projects/2026-06-29-搭上AI这条船/slides/s23/s23-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s23/s23-v1-prompt.md
@ -0,0 +1,37 @@
+# s23-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false,"base_url":"https://api.openai.com/v1","model":"gpt-image-2","env_flag_value":"(unset)"}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\poster-and-campaigns\campaign-kv.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: AGENTS.md, formal production handoff, project.md, execution-plan.md, slides.md, visual-system.md, style-prompt-master.md, motif-registry.md, P1-P24 skeleton, task-plan.md, production-run-status.md.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted customer-judgment style sample, `slides/s16/s16-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual matching Image 1 and Image 2: deep navy-black AI maritime operation map, cyan route glow, compass lines, restrained gold milestone highlights, crisp readable Chinese typography.
+
+Page title at top: “一个月训练：成为 AI 增强型客户经营人员”.
+
+Main visual: a four-week training route map across a dark AI sea chart, starting from “新船员训练” and ending at “AI增强型客户经营”. Four glowing route nodes:
+1. “第一周：和 AI 交朋友”
+2. “第二周：客户沟通”
+3. “第三周：边界规则”
+4. “第四周：模拟认证”
+
+Large central judgment: “一个月后，会用 AI 做准备、复盘和客户判断”.
+
+Bottom sentence: “不是培养技术专家，是培养 AI 增强型商务人员”.
+
+Mood: immediate action after kickoff, clear training route, confident and practical. Avoid course syllabus, school homework cartoon, formal certificate poster, dense schedule table, customer proposal, government report, SaaS dashboard, cyberpunk poster, robots, real people. Use short legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/s24/s24-v1-brief.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s24/s24-v1-brief.md
@ -0,0 +1,37 @@
+# s24-v1 Brief
+
+## Execution Card
+
+- Unit: s24
+- Source page: P24
+- Title: 第一周目标：写出自己的 AI 训练目标
+- Family: Training / Action
+- Motif: 船员训练卡
+- Running mode: prompt and generate
+- Review mode: none
+- Version: s24-v1
+- Overwrite allowed: no
+
+## Source Judgment
+
+会后第一件事：每个人形成自己的第一周 AI 训练目标。
+
+## Visual Direction
+
+Use a crew first-week training card, derived from the boarding-ticket motif. It should look actionable and fillable, but not like school homework or an administrative form.
+
+## Text To Keep Short
+
+- 第一周目标：写出自己的 AI 训练目标
+- 第一周 AI 训练目标
+- 解决什么问题
+- 每天何时用
+- 会前准备
+- 会后复盘
+- 一周成果
+- 我的 AI 工作流说明书
+- 先从自己开始，把 AI 用起来
+
+## Avoid
+
+No school test paper, no rigid form, no certificate poster, no empty slogan page.
--- a/projects/2026-06-29-搭上AI这条船/slides/s24/s24-v1-design.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s24/s24-v1-design.md
@ -0,0 +1,27 @@
+# s24-v1 Design
+
+## 页面定位
+
+- 页面：P24 / s24
+- 标题：第一周目标：写出自己的 AI 训练目标
+- 页面家族：Training / Action
+- 核心结论：会后第一件事：每个人形成自己的第一周 AI 训练目标。
+- 核心母题：船员训练卡
+
+## 视觉概念
+
+最后一页要直接落到行动。画面中心是一张“船员第一周训练卡”，像从船票升级而来的行动卡。卡片上保留五个问题，但压缩成短标签，旁边是提交成果“我的 AI 工作流说明书”。
+
+视觉上既像可填写的卡，也像 AI 航线上的第一张行动票。
+
+## 文字与布局
+
+- 顶部主标题：第一周目标：写出自己的 AI 训练目标
+- 中央卡片：第一周 AI 训练目标
+- 五问短标签：解决什么问题 / 每天何时用 / 会前准备 / 会后复盘 / 一周成果
+- 升级方向：我的 AI 工作流说明书
+- 收束金句：先从自己开始，把 AI 用起来
+
+## 风格约束
+
+保持启动会收束感，不做口号页。避免作业本、学校考试、行政表单和过度正式的证书。
--- a/projects/2026-06-29-搭上AI这条船/slides/s24/s24-v1-prompt.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/s24/s24-v1-prompt.md
@ -0,0 +1,38 @@
+# s24-v1 Prompt
+
+## gpt-image-2 Workflow Record
+
+- Skill mode result: `{"mode":"B-or-C","recommendation":"host-or-advisor","garden_mode_enabled":false,"has_api_key":false,"base_url":"https://api.openai.com/v1","model":"gpt-image-2","env_flag_value":"(unset)"}`
+- Effective route: Mode B, host-native image generation.
+- Garden expected: no.
+- Reference/template files used:
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\poster-and-campaigns\campaign-kv.md`
+  - `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
+- Source files used: AGENTS.md, formal production handoff, project.md, execution-plan.md, slides.md, visual-system.md, style-prompt-master.md, motif-registry.md, P1-P24 skeleton, task-plan.md, production-run-status.md.
+- Visual calibration references loaded:
+  - Image 1: accepted content-page style sample, `slides/s02/s02-v1-image.png`
+  - Image 2: accepted customer-judgment style sample, `slides/s16/s16-v1-image.png`
+
+## Final Image Prompt
+
+Create a 16:9 internal kickoff PPT full-page visual matching Image 1 and Image 2: deep navy-black AI maritime operation map, boarding-ticket motif, cyan route glow, restrained gold action highlights, compass grid, crisp readable Chinese typography.
+
+Page title at top: “第一周目标：写出自己的 AI 训练目标”.
+
+Main visual: a large crew first-week training card, like a boarding pass upgraded into an action worksheet. Card title: “第一周 AI 训练目标”. On the card show five short fill-in prompts as clean rows:
+1. “解决什么问题”
+2. “每天何时用”
+3. “会前准备”
+4. “会后复盘”
+5. “一周成果”
+
+Add a gold upgrade tag on the card: “我的 AI 工作流说明书”.
+
+Bottom closing sentence: “先从自己开始，把 AI 用起来”.
+
+Mood: final action page, concrete and motivating, immediately usable after the meeting. Avoid school test paper, administrative form, certificate poster, empty slogan page, customer proposal, government report, SaaS dashboard, cyberpunk poster, cartoon ship, robots, real people. Use short legible Chinese labels only.
--- a/projects/2026-06-29-搭上AI这条船/slides/slides.md
+++ b/projects/2026-06-29-搭上AI这条船/slides/slides.md
@ -0,0 +1,51 @@
+# Slides
+
+## Status
+
+- Project: `2026-06-29-搭上AI这条船`
+- Unit type: PPT page / slide.
+- Source page notation: P01-P24.
+- Local artifact notation for generated work: `s01`, `s02`, ...
+- Current status: P01/P02 samples confirmed by user; s03-s24 production-pass.
+- Write policy: append-only; overwrite allowed: no.
+
+## Source Authority
+
+Page order, page title, and core conclusion come from:
+
+```text
+../intake/阶段5：P1-P24 单页设计骨架.md
+```
+
+## Page Status Table
+
+| Unit | Source Page | Title | Page Family | Status | Design | Prompt | Image | Notes |
+| --- | --- | --- | --- | --- | --- | --- | --- | --- |
+| s01 | P01 | 搭上 AI 这条船 | Opening / Consensus | user-confirmed pass | `slides/s01/s01-v1-design.md`; `slides/s01/s01-v1-brief.md` | `slides/s01/s01-v1-prompt.md` | `slides/s01/s01-v1-image.png` | Cover and public-style sample / AI 航线 / 船票 |
+| s02 | P02 | 今天这场会，不是再动员一次 | Opening / Consensus | user-confirmed pass | `slides/s02/s02-v1-design.md`; `slides/s02/s02-v1-brief.md` | `slides/s02/s02-v1-prompt.md` | `slides/s02/s02-v1-image.png` | First content-page sample / Not-Is consensus page |
+| s03 | P03 | 我们要建的，不是传统销售团队 | Opening / Consensus | production-pass | `slides/s03/s03-v1-design.md`; `slides/s03/s03-v1-brief.md` | `slides/s03/s03-v1-prompt.md` | `slides/s03/s03-v1-image.png` | Radar-centered team identity |
+| s04 | P04 | 为什么现在必须变 | Pressure / Role Transition | production-pass | `slides/s04/s04-v1-design.md`; `slides/s04/s04-v1-brief.md` | `slides/s04/s04-v1-prompt.md` | `slides/s04/s04-v1-image.png` | Old route pressure / new AI route |
+| s05 | P05 | 我这半年打通了什么 | Pressure / Role Transition | production-pass | `slides/s05/s05-v1-design.md`; `slides/s05/s05-v1-brief.md` | `slides/s05/s05-v1-prompt.md` | `slides/s05/s05-v1-image.png` | Product / method / training base |
+| s06 | P06 | 我的角色已经变了 | Pressure / Role Transition | production-pass | `slides/s06/s06-v1-design.md`; `slides/s06/s06-v1-brief.md` | `slides/s06/s06-v1-prompt.md` | `slides/s06/s06-v1-image.png` | Overloaded center to expert node |
+| s07 | P07 | 过去靠我一个人顶，未来靠体系一起跑 | Pressure / Role Transition | production-pass | `slides/s07/s07-v1-design.md`; `slides/s07/s07-v1-brief.md` | `slides/s07/s07-v1-prompt.md` | `slides/s07/s07-v1-image.png` | Single engine to multi-engine system |
+| s08 | P08 | 为什么不是传统销售 | Confidence / Agentic Entry | production-pass | `slides/s08/s08-v1-design.md`; `slides/s08/s08-v1-brief.md` | `slides/s08/s08-v1-prompt.md` | `slides/s08/s08-v1-image.png` | Traditional sales vs customer journey |
+| s09 | P09 | 为什么这次找你们 | Confidence / Agentic Entry | production-pass | `slides/s09/s09-v1-design.md`; `slides/s09/s09-v1-brief.md` | `slides/s09/s09-v1-prompt.md` | `slides/s09/s09-v1-image.png` | 识人 / 识局 / 识势 |
+| s10 | P10 | Agentic 技术平权：外行也能上船 | Confidence / Agentic Entry | production-pass | `slides/s10/s10-v1-design.md`; `slides/s10/s10-v1-brief.md` | `slides/s10/s10-v1-prompt.md` | `slides/s10/s10-v1-image.png` | Boarding ramp / lowered technical wall |
+| s11 | P11 | 传统软件工程师和 Agentic 工程师，是两个物种 | Confidence / Agentic Entry | production-pass | `slides/s11/s11-v1-design.md`; `slides/s11/s11-v1-brief.md` | `slides/s11/s11-v1-prompt.md` | `slides/s11/s11-v1-image.png` | Capability-system contrast |
+| s12 | P12 | 第一阶段要求：不是懂 AI，而是会带着 AI 工作 | Confidence / Agentic Entry | production-pass | `slides/s12/s12-v1-design.md`; `slides/s12/s12-v1-brief.md` | `slides/s12/s12-v1-prompt.md` | `slides/s12/s12-v1-image.png` | AI work habit loop |
+| s13 | P13 | 我们卖的不是 AI 产品，是一条企业智能化路径 | Business Path / Customer Judgment | production-pass | `slides/s13/s13-v1-design.md`; `slides/s13/s13-v1-brief.md` | `slides/s13/s13-v1-prompt.md` | `slides/s13/s13-v1-image.png` | 想用 AI -> 用好 AI route |
+| s14 | P14 | 三级服务路径：培训 → 顾问 → 系统 | Business Path / Customer Judgment | production-pass | `slides/s14/s14-v1-design.md`; `slides/s14/s14-v1-brief.md` | `slides/s14/s14-v1-prompt.md` | `slides/s14/s14-v1-image.png` | Three-stage route |
+| s15 | P15 | 培训可以宽，顾问要筛，系统必须严 | Business Path / Customer Judgment | production-pass | `slides/s15/s15-v1-design.md`; `slides/s15/s15-v1-brief.md` | `slides/s15/s15-v1-prompt.md` | `slides/s15/s15-v1-image.png` | Funnel / lock gate |
+| s16 | P16 | 客户适配判断模型 | Business Path / Customer Judgment | production-pass | `slides/s16/s16-v1-design.md`; `slides/s16/s16-v1-brief.md` | `slides/s16/s16-v1-prompt.md` | `slides/s16/s16-v1-image.png` | Four-direction compass |
+| s17 | P17 | 双轨体系：先探索价值，再工程兑现 | Business Path / Customer Judgment | production-pass | `slides/s17/s17-v1-design.md`; `slides/s17/s17-v1-brief.md` | `slides/s17/s17-v1-prompt.md` | `slides/s17/s17-v1-image.png` | Double-track exploration / engineering route |
+| s18 | P18 | 新团队协作结构 | Team Structure / Boundaries | production-pass | `slides/s18/s18-v1-design.md`; `slides/s18/s18-v1-brief.md` | `slides/s18/s18-v1-prompt.md` | `slides/s18/s18-v1-image.png` | Four-role ship-compartment operation map |
+| s19 | P19 | 我们是帮客户上船的人 | Team Structure / Boundaries | production-pass | `slides/s19/s19-v1-design.md`; `slides/s19/s19-v1-brief.md` | `slides/s19/s19-v1-prompt.md` | `slides/s19/s19-v1-image.png` | Five role icons around boarding path |
+| s20 | P20 | 魏征机制：你们也是我的现实校准器 | Team Structure / Boundaries | production-pass | `slides/s20/s20-v1-design.md`; `slides/s20/s20-v1-brief.md` | `slides/s20/s20-v1-prompt.md` | `slides/s20/s20-v1-image.png` | Compass correction / reality calibration |
+| s21 | P21 | L0-L4 成长路径 | Team Structure / Boundaries | production-pass | `slides/s21/s21-v1-design.md`; `slides/s21/s21-v1-brief.md` | `slides/s21/s21-v1-prompt.md` | `slides/s21/s21-v1-image.png` | Ladder / navigation license |
+| s22 | P22 | 今天先记住三条 | Team Structure / Boundaries | production-pass | `slides/s22/s22-v1-design.md`; `slides/s22/s22-v1-brief.md` | `slides/s22/s22-v1-prompt.md` | `slides/s22/s22-v1-image.png` | Three safety gates |
+| s23 | P23 | 一个月训练：成为 AI 增强型客户经营人员 | Training / Action | production-pass | `slides/s23/s23-v1-design.md`; `slides/s23/s23-v1-brief.md` | `slides/s23/s23-v1-prompt.md` | `slides/s23/s23-v1-image.png` | Four-week training route |
+| s24 | P24 | 第一周目标：写出自己的 AI 训练目标 | Training / Action | production-pass | `slides/s24/s24-v1-design.md`; `slides/s24/s24-v1-brief.md` | `slides/s24/s24-v1-prompt.md` | `slides/s24/s24-v1-image.png` | Crew first-week training card |
+
+## Next Action
+
+All planned pages s01-s24 now have production-pass images. Use `slides/s01/s01-v1-image.png` and `slides/s02/s02-v1-image.png` as accepted style samples when future repair or append-only retries are requested.
--- a/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/production-run-status.md
+++ b/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/production-run-status.md
@ -0,0 +1,41 @@
+# Production Run Status
+
+## Status
+
+- Project: `2026-06-29-搭上AI这条船`
+- Lifecycle phase: formal production
+- Line: production-light
+- Running mode: prompt and generate
+- Review mode: none
+- Package size: 7, with final package `s17-s24` expanded to 8 by user instruction
+- User confirmation: clean callback may continue automatically
+- Current status: complete
+
+## Scope
+
+- Total target units: `s03-s24`
+- Completed units: `s03-s24`
+- Active child session: none
+- Next cursor: complete
+
+## Callback Target
+
+- Parent thread: `019f1359-47c9-79e2-b7d4-3d461eb04e28`
+- Callback mode: child sends concise callback as final action
+
+## Latest Callback
+
+- Status: complete
+- Topology: direct-child
+- Resolved versions: `s17-v1; s18-v1; s19-v1; s20-v1; s21-v1; s22-v1; s23-v1; s24-v1`
+- Handoff path: `projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/session-03/child-session-handoff.md`
+- Image paths: `projects/2026-06-29-搭上AI这条船/slides/s17/s17-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s18/s18-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s19/s19-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s20/s20-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s21/s21-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s22/s22-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s23/s23-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s24/s24-v1-image.png`
+- Problems: none blocking; `s20` first generation discarded for title typo and project image is the retry
+- Needs user: none
+
+## Stop Conditions
+
+- Artifact collision on any target `sNN-vN-*` path.
+- Generated pages drift away from the accepted P01/P02 style.
+- Text becomes too broken for internal meeting use.
+- The deck drifts into customer proposal, government report, SaaS dashboard, or cyberpunk poster style.
--- a/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/session-01/child-session-handoff.md
+++ b/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/session-01/child-session-handoff.md
@ -0,0 +1,11 @@
+# Child Session 01 Handoff
+
+- Status: complete
+- Topology: direct-child
+- Callback: ready-to-send-as-final-action
+- Review mode: none
+- Resolved versions: s03-v1; s04-v1; s05-v1; s06-v1; s07-v1; s08-v1; s09-v1
+- Image paths: `projects/2026-06-29-搭上AI这条船/slides/s03/s03-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s04/s04-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s05/s05-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s06/s06-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s07/s07-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s08/s08-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s09/s09-v1-image.png`
+- Problems: none
+- Needs user: none
+- Parent summary: Package 1 s03-s09 completed in production-light with direct child topology. Wrote v1 design, brief, prompt, and image for each assigned slide; updated `slides/slides.md` rows to production-pass.
--- a/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/session-02/child-session-handoff.md
+++ b/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/session-02/child-session-handoff.md
@ -0,0 +1,11 @@
+# Child Session 02 Handoff
+
+- Status: complete
+- Topology: direct-child
+- Callback: ready-to-send-as-final-action
+- Review mode: none
+- Resolved versions: `s10-v1`; `s11-v1`; `s12-v1`; `s13-v1`; `s14-v1`; `s15-v1`; `s16-v1`
+- Image paths: `projects/2026-06-29-搭上AI这条船/slides/s10/s10-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s11/s11-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s12/s12-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s13/s13-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s14/s14-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s15/s15-v1-image.png`; `projects/2026-06-29-搭上AI这条船/slides/s16/s16-v1-image.png`
+- Problems: none
+- Needs user: none
+- Parent summary: Completed formal production package 2 for `s10-s16` under production-light, direct-child, review none. Wrote design, brief, prompt, generated image, and updated `slides/slides.md` rows for all assigned pages.
--- a/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/session-03/child-session-handoff.md
+++ b/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/session-03/child-session-handoff.md
@ -0,0 +1,11 @@
+# Child Session 03 Handoff
+
+- Status: complete
+- Topology: direct-child
+- Callback: ready-to-send-as-final-action
+- Review mode: none
+- Resolved versions: s17-v1; s18-v1; s19-v1; s20-v1; s21-v1; s22-v1; s23-v1; s24-v1
+- Image paths: projects/2026-06-29-搭上AI这条船/slides/s17/s17-v1-image.png; projects/2026-06-29-搭上AI这条船/slides/s18/s18-v1-image.png; projects/2026-06-29-搭上AI这条船/slides/s19/s19-v1-image.png; projects/2026-06-29-搭上AI这条船/slides/s20/s20-v1-image.png; projects/2026-06-29-搭上AI这条船/slides/s21/s21-v1-image.png; projects/2026-06-29-搭上AI这条船/slides/s22/s22-v1-image.png; projects/2026-06-29-搭上AI这条船/slides/s23/s23-v1-image.png; projects/2026-06-29-搭上AI这条船/slides/s24/s24-v1-image.png
+- Problems: none blocking; s20 first generation was discarded for title typo and the copied image is the retry
+- Needs user: none
+- Parent summary: package 3 s17-s24 completed under production-light, direct-child topology, review none; design/brief/prompt/image files written append-only as v1 and slides/slides.md rows updated to production-pass.
--- a/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/task-log.md
+++ b/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/task-log.md
@ -0,0 +1,24 @@
+# Task Log
+
+## Status
+
+- Current status: complete.
+
+## Dispatch / Callback Receipts
+
+- 2026-06-29 20:34 dispatch: session-01 `019f1360-8d74-7cf0-9ce3-59305fa7b43f`, package 1 `s03-s09`, line `production-light`, review `none`, callback target `019f1359-47c9-79e2-b7d4-3d461eb04e28`.
+- 2026-06-29 20:57 callback: complete, direct-child, versions `s03-v1; s04-v1; s05-v1; s06-v1; s07-v1; s08-v1; s09-v1`, handoff `projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/session-01/child-session-handoff.md`, needs_user none.
+- 2026-06-29 20:57 dispatch: session-02 `019f1375-ffe8-78b0-86b7-63c1429970bf`, package 2 `s10-s16`, line `production-light`, review `none`, callback target `019f1359-47c9-79e2-b7d4-3d461eb04e28`.
+- 2026-06-29 21:23 callback: complete, direct-child, versions `s10-v1; s11-v1; s12-v1; s13-v1; s14-v1; s15-v1; s16-v1`, handoff `projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/session-02/child-session-handoff.md`, needs_user none.
+- 2026-06-29 21:23 dispatch: session-03 `019f138d-5075-7c50-b23c-0f26954970f6`, package 3 `s17-s24`, line `production-light`, review `none`, callback target `019f1359-47c9-79e2-b7d4-3d461eb04e28`.
+- 2026-06-29 21:49 callback: complete, direct-child, versions `s17-v1; s18-v1; s19-v1; s20-v1; s21-v1; s22-v1; s23-v1; s24-v1`, handoff `projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/session-03/child-session-handoff.md`, needs_user none, note `s20` first generation discarded for title typo and project image is the retry.
+
+## Slide Status Summary
+
+- Target scope: `s03-s24`.
+- Completed in this production run: `s03-s24`.
+- Next package: none.
+
+## Open Decisions
+
+- None.
--- a/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/task-plan.md
+++ b/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-formal-production/task-plan.md
@ -0,0 +1,33 @@
+# 2026-06-29 Formal Production
+
+## Active Scope
+
+- Line: production-light
+- Worker profile: PPT-page-production-lite
+- Running mode: prompt and generate
+- Review mode: none
+- Slides: s03-s24
+- Package split: s03-s09, s10-s16, s17-s24
+- Child session: session-03 complete / `019f138d-5075-7c50-b23c-0f26954970f6` / `子会话：AI船 s17-s24`
+- Callback target: 019f1359-47c9-79e2-b7d4-3d461eb04e28
+- Write policy: append-only; overwrite allowed: no
+- Status: complete
+
+## Context To Use
+
+- `AGENTS.md`
+- `handoff/2026-06-29-ai-ship-formal-production-handoff.md`
+- `project.md`
+- `execution-plan.md`
+- `slides/slides.md`
+- `visual-system/visual-system.md`
+- `visual-system/style/style-prompt-master.md`
+- `visual-system/motifs/motif-registry.md`
+- `intake/阶段5：P1-P24 单页设计骨架.md`
+
+## Stop Conditions
+
+- Artifact collision on any target `sNN-vN-*` path.
+- Generated pages drift away from the accepted P01/P02 style.
+- Text becomes too broken for internal meeting use.
+- The deck drifts into customer proposal, government report, SaaS dashboard, or cyberpunk poster style.
--- a/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-visual-system-alignment/lifecycle-guard-config.yaml
+++ b/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-visual-system-alignment/lifecycle-guard-config.yaml
@ -0,0 +1,29 @@
+watched_paths:
+  - "*.md"
+status_fields:
+  - status
+  - lifecycle
+forbidden_status_values:
+  - stable
+  - final
+  - released
+  - production-ready
+warning_status_values:
+  - accepted
+  - confirmed
+required_evidence_markers:
+  - user_decision
+  - evidence
+approved_phrases:
+  - owner-approved
+  - ccra-approved
+  - production-ready
+  - final
+  - released
+forbidden_phrases:
+  - production-ready
+allowed_contexts:
+  - "Do not"
+  - "do not"
+  - "not an"
+  - "example:"
--- a/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-visual-system-alignment/lifecycle-guard-output/lifecycle-status-guard-scan.json
+++ b/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-visual-system-alignment/lifecycle-guard-output/lifecycle-status-guard-scan.json
@ -0,0 +1,24 @@
+{
+  "scan_root": "C:\\Users\\wangq\\Documents\\Codex\\video-workbench\\projects\\2026-06-29-搭上AI这条船\\handoff",
+  "config_path": "C:\\Users\\wangq\\Documents\\Codex\\video-workbench\\projects\\2026-06-29-搭上AI这条船\\tasks\\2026-06-29-visual-system-alignment\\lifecycle-guard-config.yaml",
+  "files_scanned": [
+    "2026-06-29-ai-ship-formal-production-handoff.md"
+  ],
+  "possible_overclaims": [],
+  "field_level_findings": [],
+  "phrase_level_findings": [],
+  "missing_evidence_markers": [],
+  "warnings": [],
+  "blocking_errors": [],
+  "observations": [],
+  "machine_readable_summary": {
+    "status": "PASS",
+    "files_scanned_count": 1,
+    "field_finding_count": 0,
+    "phrase_finding_count": 0,
+    "missing_evidence_count": 0,
+    "warning_count": 0,
+    "blocking_count": 0,
+    "observation_count": 0
+  }
+}
--- a/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-visual-system-alignment/lifecycle-guard-output/lifecycle-status-guard-scan.md
+++ b/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-visual-system-alignment/lifecycle-guard-output/lifecycle-status-guard-scan.md
@ -0,0 +1,25 @@
+# Lifecycle Status Guard Scan
+
+## Summary
+
+- Status: `PASS`
+- Files scanned: 1
+- Blocking findings: 0
+- Warnings: 0
+- Evidence-present observations: 0
+
+## Blocking Findings
+
+- None
+
+## Warnings
+
+- None
+
+## Observations
+
+- None
+
+## Files Scanned
+
+- `2026-06-29-ai-ship-formal-production-handoff.md`
--- a/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-visual-system-alignment/task-log.md
+++ b/projects/2026-06-29-搭上AI这条船/tasks/2026-06-29-visual-system-alignment/task-log.md
@ -0,0 +1,20 @@
+# Task Log
+
+## Status
+
+- Current status: P01/P02 samples user-confirmed; formal production handoff ready.
+
+## Receipts
+
+- 2026-06-29 setup: user chose full-page visual-image-led PPT.
+- 2026-06-29 setup: source charter path recorded as fact-check backup only.
+- 2026-06-29 setup: visual-system files initialized; no image generation authorized.
+- 2026-06-29 alignment: user approved unified deep blue-black AI route operation-map base across every page, with only light motif changes by page family.
+- 2026-06-29 alignment: user approved P01 as both cover and public-style sample, then P02 as content-page sample; if both pass, move to production.
+- 2026-06-29 design: wrote P01 and P02 design drafts for user review: `slides/s01/s01-v1-design.md`, `slides/s02/s02-v1-design.md`.
+- 2026-06-29 generation: user approved both designs; main session wrote `s01-v1-brief.md`, `s01-v1-prompt.md`, `s02-v1-brief.md`, `s02-v1-prompt.md`; generated and copied images to `slides/s01/s01-v1-image.png` and `slides/s02/s02-v1-image.png`. Review files, child sessions, and completeness gates intentionally not used.
+- 2026-06-29 decision: user confirmed both generated images passed and asked for a handoff document; next new session will start formal production.
+
+## Open Decisions
+
+- New clean session loads the formal production handoff and starts from P03 / `s03`.
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
wantsong	63772412b4	docs: add remotion workflow rules	2026-07-07 18:11:45 +08:00
wantsong	31b920e819	Add deck project text records	2026-07-07 17:11:01 +08:00