diff --git a/skills/creative/baoyu-comic/PORT_NOTES.md b/skills/creative/baoyu-comic/PORT_NOTES.md
index 24b21db873..142a08d1ee 100644
--- a/skills/creative/baoyu-comic/PORT_NOTES.md
+++ b/skills/creative/baoyu-comic/PORT_NOTES.md
@@ -12,10 +12,10 @@ Ported from [JimLiu/baoyu-skills](https://github.com/JimLiu/baoyu-skills) v1.56.
 | Trigger | Slash commands / CLI flags | Natural language skill matching |
 | User config | EXTEND.md file (project/user/XDG paths) | Removed — not part of Hermes infra |
 | User prompts | `AskUserQuestion` (batched) | `clarify` tool (one question at a time) |
-| Image generation | baoyu-imagine (Bun/TypeScript) | `image_generate` tool |
+| Image generation | baoyu-imagine (Bun/TypeScript, supports `--ref`) | `image_generate` — **prompt-only**, returns a URL; no reference image input; agent must download the URL to the output directory |
+| PDF assembly | `scripts/merge-to-pdf.ts` (Bun + `pdf-lib`) | Removed — the PDF merge step is out of scope for this port; pages are delivered as PNGs only |
 | Platform support | Linux/macOS/Windows/WSL/PowerShell | Linux/macOS only |
 | File operations | Generic instructions | Hermes file tools (`write_file`, `read_file`) |
-| Runtime abstraction | `${BUN_X}` resolution | Direct `bun` invocation for `scripts/merge-to-pdf.ts` |
 
 ### Structural removals
 
@@ -23,17 +23,32 @@ Ported from [JimLiu/baoyu-skills](https://github.com/JimLiu/baoyu-skills) v1.56.
   - `first-time-setup.md` — blocking first-time setup flow for EXTEND.md
   - `preferences-schema.md` — EXTEND.md YAML schema
   - `watermark-guide.md` — watermark config (tied to EXTEND.md)
+- **`scripts/` directory** (removed entirely): upstream's `merge-to-pdf.ts` depended on `pdf-lib`, which is not declared anywhere in the Hermes repo. Rather than add a new dependency, the port drops PDF assembly and delivers per-page PNGs.
+- **Workflow Step 8 (Merge to PDF)** removed from `workflow.md`; Step 9 (Completion report) renumbered to Step 8.
 - **Workflow Step 1.1** — "Load Preferences (EXTEND.md)" section removed from `workflow.md`; steps 1.2/1.3 renumbered to 1.1/1.2.
 - **Generic "User Input Tools" and "Image Generation Tools" preambles** — SKILL.md no longer lists fallback rules for multiple possible tools; it references `clarify` and `image_generate` directly.
 
+### Image generation strategy changes
+
+`image_generate`'s schema accepts only `prompt` and `aspect_ratio` (`landscape` | `portrait` | `square`). Upstream's reference-image flow (`--ref characters.png` for character consistency, plus user-supplied refs for style/palette/scene) does not map to this tool, so the workflow was restructured:
+
+- **Character sheet** is still generated, but it is an **agent-facing** reference used when writing each page's prompt text. `image_generate` never sees it as a visual input.
+- **User-supplied reference images** are reduced to `style` / `palette` / `scene` trait extraction — traits are embedded in the prompt body; the image files themselves are kept only for provenance under `refs/`.
+- **Page prompts** now mandate that character descriptions are embedded inline (copied from `characters/characters.md`) — this is the only mechanism left to enforce cross-page character consistency.
+- **Download step** — after every `image_generate` call, the returned URL is fetched to disk (e.g., `curl -fsSL "<url>" -o <target>.png`) and verified before the workflow advances.
+
 ### SKILL.md reductions
 
 - CLI option columns (`--art`, `--tone`, `--layout`, `--aspect`, `--lang`, `--ref`, `--storyboard-only`, `--prompts-only`, `--images-only`, `--regenerate`) converted to plain-English option descriptions.
 - Preset files (`presets/*.md`) and `ohmsha-guide.md`: `` `--style X` `` / `` `--art X --tone Y` `` shorthand rewritten to `art=X, tone=Y` + natural-language references.
-- `partial-workflows.md`: per-skill slash command invocations rewritten as user-intent cues.
+- `partial-workflows.md`: per-skill slash command invocations rewritten as user-intent cues; PDF-related outputs removed.
 - `auto-selection.md`: priority order dropped the EXTEND.md tier.
 - `analysis-framework.md`: language-priority comment updated (user option → conversation → source).
 
+### File naming convention
+
+Source content pasted by the user is saved as `source-{slug}.md`, where `{slug}` is the kebab-case topic slug used for the output directory. Backups follow the same pattern with a `-backup-YYYYMMDD-HHMMSS` suffix. SKILL.md and `workflow.md` now agree on this single convention.
+
 ### What was preserved verbatim
 
 - All 6 art-style definitions (`references/art-styles/`)
@@ -41,7 +56,6 @@ Ported from [JimLiu/baoyu-skills](https://github.com/JimLiu/baoyu-skills) v1.56.
 - All 7 layout definitions (`references/layouts/`)
 - Core templates: `character-template.md`, `storyboard-template.md`, `base-prompt.md`
 - Preset bodies (only the first few intro lines adapted; special rules unchanged)
-- `scripts/merge-to-pdf.ts` (Bun-compatible on Linux/macOS)
 - Author, version, homepage attribution
 
 ## Syncing with upstream
@@ -59,3 +73,5 @@ diff <(curl -sL https://raw.githubusercontent.com/JimLiu/baoyu-skills/main/skill
 ```
 
 Art-style, tone, and layout reference files can usually be overwritten directly (they're upstream-verbatim). `SKILL.md`, `references/workflow.md`, `references/partial-workflows.md`, `references/auto-selection.md`, `references/analysis-framework.md`, `references/ohmsha-guide.md`, and `references/presets/*.md` must be manually merged since they contain Hermes-specific adaptations.
+
+If upstream adds a Hermes-compatible PDF merge step (no extra npm deps), restore `scripts/` and reintroduce Step 8 in `workflow.md`.
diff --git a/skills/creative/baoyu-comic/SKILL.md b/skills/creative/baoyu-comic/SKILL.md
index e6ec91d6fd..9c950cbac0 100644
--- a/skills/creative/baoyu-comic/SKILL.md
+++ b/skills/creative/baoyu-comic/SKILL.md
@@ -22,10 +22,10 @@ Trigger this skill when the user asks to create a knowledge/educational comic, b
 
 ## Reference Images
 
-Users may supply reference images to guide art style, palette, scene composition, or subject. This is **separate from** the auto-generated character sheet (Step 7.1) — both can coexist: user refs guide the look, the character sheet anchors recurring character identity.
+Hermes' `image_generate` tool is **prompt-only** — it accepts a text prompt and an aspect ratio, and returns an image URL. It does **NOT** accept reference images. When the user supplies a reference image, use it to **extract traits in text** that get embedded in every page prompt:
 
 **Intake**: Accept file paths when the user provides them (or pastes images in conversation).
-- File path(s) → copy to `refs/NN-ref-{slug}.{ext}` alongside the comic output
+- File path(s) → copy to `refs/NN-ref-{slug}.{ext}` alongside the comic output for provenance
 - Pasted image with no path → ask the user for the path via `clarify`, or extract style traits verbally as a text fallback
 - No reference → skip this section
 
@@ -33,9 +33,9 @@ Users may supply reference images to guide art style, palette, scene composition
 
 | Usage | Effect |
 |-------|--------|
-| `direct` | Pass the file to `image_generate` as a reference image on every page (or selected pages) |
 | `style` | Extract style traits (line treatment, texture, mood) and append to every page's prompt body |
 | `palette` | Extract hex colors and append to every page's prompt body |
+| `scene` | Extract scene composition or subject notes and append to the relevant page(s) |
 
 **Record in each page's prompt frontmatter** when refs exist:
 
@@ -43,14 +43,11 @@ Users may supply reference images to guide art style, palette, scene composition
 references:
   - ref_id: 01
     filename: 01-ref-scene.png
-    usage: direct
+    usage: style
+    traits: "muted earth tones, soft-edged ink wash, low-contrast backgrounds"
 ```
 
-**At generation time**:
-- Verify each referenced file exists on disk
-- If `usage: direct` AND `image_generate` accepts multiple reference images → pass both the character sheet (Step 7.2) and the user refs; compress images first per Step 7.1's guidance to avoid payload failures
-- If only one ref slot is available → prefer the character sheet for pages with recurring characters; embed user-ref traits in the prompt body instead
-- For `style`/`palette` usage → embed extracted traits in every page's prompt text
+Character consistency is still driven by the **character sheet workflow** (Step 7.1–7.2) below, which relies on detailed text descriptions rather than direct image references.
 
 ## Options
 
@@ -63,7 +60,7 @@ references:
 | Layout | standard (default), cinematic, dense, splash, mixed, webtoon, four-panel | Panel arrangement |
 | Aspect | 3:4 (default, portrait), 4:3 (landscape), 16:9 (widescreen) | Page aspect ratio |
 | Language | auto (default), zh, en, ja, etc. | Output language |
-| Refs | File paths | Reference images applied to every page for style / palette / scene guidance. See [Reference Images](#reference-images) above. |
+| Refs | File paths | Reference images used for style / palette trait extraction (not passed to the image model). See [Reference Images](#reference-images) above. |
 
 ### Partial Workflow Options
 
@@ -94,14 +91,6 @@ Details: [references/partial-workflows.md](references/partial-workflows.md)
 
 - **Compatibility matrix** and **content-signal → preset** table live in [references/auto-selection.md](references/auto-selection.md). Read it before recommending combinations in Step 2.
 
-## Scripts
-
-| Script | Purpose |
-|--------|---------|
-| `scripts/merge-to-pdf.ts` | Merge comic pages into PDF (runs with `bun`) |
-
-Resolve `{baseDir}` as this SKILL.md's directory; script path is `{baseDir}/scripts/merge-to-pdf.ts`.
-
 ## File Structure
 
 Output directory: `comic/{topic-slug}/`
@@ -111,14 +100,14 @@ Output directory: `comic/{topic-slug}/`
 **Contents**:
 | File | Description |
 |------|-------------|
-| `source-{slug}.{ext}` | Source files |
+| `source-{slug}.md` | Saved source content (kebab-case slug matches the output directory) |
 | `analysis.md` | Content analysis |
 | `storyboard.md` | Storyboard with panel breakdown |
 | `characters/characters.md` | Character definitions |
-| `characters/characters.png` | Character reference sheet |
+| `characters/characters.png` | Character reference sheet (downloaded from `image_generate`) |
 | `prompts/NN-{cover\|page}-[slug].md` | Generation prompts |
-| `NN-{cover\|page}-[slug].png` | Generated images |
-| `{topic-slug}.pdf` | Final merged PDF |
+| `NN-{cover\|page}-[slug].png` | Generated images (downloaded from `image_generate`) |
+| `refs/NN-ref-{slug}.{ext}` | User-supplied reference images (optional, for provenance) |
 
 ## Language Handling
 
@@ -151,22 +140,21 @@ Comic Progress:
 - [ ] Step 6: Review prompts (conditional)
 - [ ] Step 7: Generate images
   - [ ] 7.1 Generate character sheet (if needed) → characters/characters.png
-  - [ ] 7.2 Generate pages (with character ref if sheet exists)
-- [ ] Step 8: Merge to PDF
-- [ ] Step 9: Completion report
+  - [ ] 7.2 Generate pages (with character descriptions embedded in prompt)
+- [ ] Step 8: Completion report
 ```
 
 ### Flow
 
 ```
-Input → Analyze → [Check Existing?] → [Confirm: Style + Reviews] → Storyboard → [Review?] → Prompts → [Review?] → Images → PDF → Complete
+Input → Analyze → [Check Existing?] → [Confirm: Style + Reviews] → Storyboard → [Review?] → Prompts → [Review?] → Images → Complete
 ```
 
 ### Step Summary
 
 | Step | Action | Key Output |
 |------|--------|------------|
-| 1.1 | Analyze content | `analysis.md` |
+| 1.1 | Analyze content | `analysis.md`, `source-{slug}.md` |
 | 1.2 | Check existing directory | Handle conflicts |
 | 2 | Confirm style, focus, audience, reviews | User preferences |
 | 3 | Generate storyboard + characters | `storyboard.md`, `characters/` |
@@ -174,9 +162,8 @@ Input → Analyze → [Check Existing?] → [Confirm: Style + Reviews] → Story
 | 5 | Generate prompts | `prompts/*.md` |
 | 6 | Review prompts (if requested) | User approval |
 | 7.1 | Generate character sheet (if needed) | `characters/characters.png` |
-| 7.2 | Generate pages (with character ref if available) | `*.png` files |
-| 8 | Merge to PDF | `{slug}.pdf` |
-| 9 | Completion report | Summary |
+| 7.2 | Generate pages | `*.png` files |
+| 8 | Completion report | Summary |
 
 ### User Questions
 
@@ -184,11 +171,11 @@ Use the `clarify` tool to confirm options. Since `clarify` handles one question
 
 ### Step 7: Image Generation
 
-Use Hermes' built-in `image_generate` tool for all image rendering.
+Use Hermes' built-in `image_generate` tool for all image rendering. Its schema accepts only `prompt` and `aspect_ratio` (`landscape` | `portrait` | `square`); it **returns a URL**, not a local file. Every generated page or character sheet must therefore be downloaded to the output directory.
 
 **Prompt file requirement (hard)**: write each image's full, final prompt to a standalone file under `prompts/` (naming: `NN-{type}-[slug].md`) BEFORE calling `image_generate`. The prompt file is the reproducibility record.
 
-**Aspect ratio mapping** — `image_generate` supports `landscape`, `portrait`, and `square`. Map as follows:
+**Aspect ratio mapping** — the storyboard's `aspect_ratio` field maps to `image_generate`'s format as follows:
 
 | Storyboard ratio | `image_generate` format |
 |------------------|-------------------------|
@@ -196,19 +183,21 @@ Use Hermes' built-in `image_generate` tool for all image rendering.
 | `4:3`, `16:9`, `3:2` | `landscape` |
 | `1:1` | `square` |
 
-**7.1 Character sheet** — generate it (to `characters/characters.png`, aspect `landscape`) when the comic is multi-page with recurring characters. Skip for simple presets (e.g., four-panel minimalist) or single-page comics. Compress to JPEG before using as a reference (`sips -s format jpeg -s formatOptions 80 …` on macOS, `pngquant --quality=65-80 …` on Linux) to avoid payload failures. The prompt file at `characters/characters.md` must exist before invoking `image_generate`.
+**Download step** — after every `image_generate` call:
+1. Read the URL from the tool result
+2. Fetch the image bytes (e.g., `curl -fsSL "<url>" -o <target>.png`)
+3. Verify the file exists and is non-empty before proceeding to the next page
 
-**7.2 Pages** — each page's prompt MUST already be at `prompts/NN-{cover|page}-[slug].md` before invoking `image_generate`. Strategy depends on the character sheet:
+**7.1 Character sheet** — generate it (to `characters/characters.png`, aspect `landscape`) when the comic is multi-page with recurring characters. Skip for simple presets (e.g., four-panel minimalist) or single-page comics. The prompt file at `characters/characters.md` must exist before invoking `image_generate`. After download, the character sheet is consumed **for the agent's own reference** when writing each page's prompt text — Hermes' `image_generate` cannot accept it as a visual input.
 
-| Character sheet | `image_generate` reference support | Strategy |
-|-----------------|------------------------------------|----------|
-| Exists | Supported | Pass sheet as reference image on every page |
-| Exists | Not supported | Prepend character descriptions to every prompt file |
-| Skipped | — | All descriptions inline in prompt |
+**7.2 Pages** — each page's prompt MUST already be at `prompts/NN-{cover|page}-[slug].md` before invoking `image_generate`. Because `image_generate` is prompt-only, character consistency is enforced by **embedding character descriptions in every prompt**:
 
-**Backup rule**: existing `prompts/…md` and `…png` files → rename with `-backup-YYYYMMDD-HHMMSS` suffix (use `write_file` / standard shell rename) before regenerating. Aspect ratio from storyboard (default `3:4`; preset may override).
+| Character sheet | Strategy |
+|-----------------|----------|
+| Exists | Prepend relevant character descriptions (from `characters/characters.md`) to every page prompt |
+| Skipped | Prompt file already contains all descriptions inline |
 
-**Reference failure recovery**: compress sheet → retry → still fails → drop the reference and embed character descriptions in the prompt text.
+**Backup rule**: existing `prompts/…md` and `…png` files → rename with `-backup-YYYYMMDD-HHMMSS` suffix before regenerating.
 
 Full step-by-step workflow (analysis, storyboard, review gates, regeneration variants): [references/workflow.md](references/workflow.md).
 
@@ -235,18 +224,18 @@ Full step-by-step workflow (analysis, storyboard, review gates, regeneration var
 
 | Action | Steps |
 |--------|-------|
-| **Edit** | **Update prompt file FIRST** → regenerate image → regenerate PDF |
-| **Add** | Create prompt at position → generate with character ref → renumber subsequent → update storyboard → regenerate PDF |
-| **Delete** | Remove files → renumber subsequent → update storyboard → regenerate PDF |
+| **Edit** | **Update prompt file FIRST** → regenerate image → download new PNG |
+| **Add** | Create prompt at position → generate with character descriptions embedded → renumber subsequent → update storyboard |
+| **Delete** | Remove files → renumber subsequent → update storyboard |
 
 **IMPORTANT**: When updating pages, ALWAYS update the prompt file (`prompts/NN-{cover|page}-[slug].md`) FIRST before regenerating. This ensures changes are documented and reproducible.
 
 ## Pitfalls
 
 - Image generation: 10-30 seconds per page; auto-retry once on failure
+- **Always download** the URL returned by `image_generate` to a local PNG — downstream tooling (and the user's review) expects files in the output directory, not ephemeral URLs
 - Use stylized alternatives for sensitive public figures
 - **Step 2 confirmation required** - do not skip
 - **Steps 4/6 conditional** - only if user requested in Step 2
-- **Step 7.1 character sheet** - recommended for multi-page comics, optional for simple presets
-- **Step 7.2 character reference** - pass sheet as reference if it exists; compress/convert on failure; fall back to prompt-only
+- **Step 7.1 character sheet** - recommended for multi-page comics, optional for simple presets. It is an **agent-facing** reference used to write consistent page prompts; `image_generate` does not accept it as a visual input
 - **Strip secrets** — scan source content for API keys, tokens, or credentials before writing any output file
diff --git a/skills/creative/baoyu-comic/references/partial-workflows.md b/skills/creative/baoyu-comic/references/partial-workflows.md
index 816aa61e09..749b5ac7b2 100644
--- a/skills/creative/baoyu-comic/references/partial-workflows.md
+++ b/skills/creative/baoyu-comic/references/partial-workflows.md
@@ -8,8 +8,8 @@ Options to run specific parts of the workflow. Trigger these via natural languag
 |--------|----------------|--------|
 | Storyboard only | 1-3 | `storyboard.md` + `characters/` |
 | Prompts only | 1-5 | + `prompts/*.md` |
-| Images only | 7-9 | + images + PDF |
-| Regenerate N | 7 (partial) | Specific page(s) + PDF |
+| Images only | 7-8 | + images |
+| Regenerate N | 7 (partial) | Specific page(s) |
 
 ---
 
@@ -60,7 +60,7 @@ Generate images from existing prompts (starts at Step 7).
 
 **User cue**: "generate images from existing prompts", "run the images now" (pointing at an existing `comic/topic-slug/` directory).
 
-**Workflow**: Skip to Step 7, then 8-9
+**Workflow**: Skip to Step 7, then 8
 
 **Prerequisites** (must exist in directory):
 - `prompts/` directory with page prompt files
@@ -70,7 +70,6 @@ Generate images from existing prompts (starts at Step 7).
 **Output**:
 - `characters/characters.png` (if not exists)
 - `NN-{cover|page}-[slug].png` images
-- `{topic-slug}.pdf`
 
 **Use case**: Re-generate images after editing prompts. Useful for:
 - Recovering from failed image generation
@@ -88,15 +87,14 @@ Regenerate specific pages only.
 **Workflow**:
 1. Read existing prompts for specified pages
 2. Regenerate images only for those pages via `image_generate`
-3. Regenerate PDF
+3. Download each returned URL and overwrite the existing PNG
 
 **Prerequisites** (must exist):
 - `prompts/NN-{cover|page}-[slug].md` for specified pages
-- `characters/characters.png` (for reference, if it was used originally)
+- `characters/characters.md` (for agent-side consistency checks, if it was used originally)
 
 **Output**:
 - Regenerated `NN-{cover|page}-[slug].png` for specified pages
-- Updated `{topic-slug}.pdf`
 
 **Use case**: Fix specific pages without regenerating entire comic. Useful for:
 - Fixing a single problematic page
diff --git a/skills/creative/baoyu-comic/references/workflow.md b/skills/creative/baoyu-comic/references/workflow.md
index a2de541c96..e8c811beb5 100644
--- a/skills/creative/baoyu-comic/references/workflow.md
+++ b/skills/creative/baoyu-comic/references/workflow.md
@@ -19,14 +19,13 @@ Comic Progress:
 - [ ] Step 7: Generate images
   - [ ] 7.1 Character sheet (if needed)
   - [ ] 7.2 Generate pages
-- [ ] Step 8: Merge to PDF
-- [ ] Step 9: Completion report
+- [ ] Step 8: Completion report
 ```
 
 ## Flow Diagram
 
 ```
-Input → Analyze → [Check Existing?] → [Confirm: Style + Reviews] → Storyboard → [Review Outline?] → Prompts → [Review Prompts?] → Images → PDF → Complete
+Input → Analyze → [Check Existing?] → [Confirm: Style + Reviews] → Storyboard → [Review Outline?] → Prompts → [Review Prompts?] → Images → Complete
 ```
 
 ---
@@ -40,8 +39,8 @@ Read source content, save it if needed, and perform deep analysis.
 **Actions**:
 1. **Save source content** (if not already a file):
    - If user provides a file path: use as-is
-   - If user pastes content: save to `source.md` in target directory using `write_file`
-   - **Backup rule**: If `source.md` exists, rename to `source-backup-YYYYMMDD-HHMMSS.md`
+   - If user pastes content: save to `source-{slug}.md` in the target directory using `write_file`, where `{slug}` is the kebab-case topic slug used for the output directory
+   - **Backup rule**: If `source-{slug}.md` already exists, rename it to `source-{slug}-backup-YYYYMMDD-HHMMSS.md` before writing
 2. Read source content
 3. **Deep analysis** following `analysis-framework.md`:
    - Target audience identification
@@ -246,7 +245,7 @@ Create image generation prompts for all pages.
 
 **For each page (cover + pages)**:
 1. Create prompt following art style + tone guidelines
-2. Include character visual descriptions for consistency
+2. **Embed character descriptions** inline (copy relevant traits from `characters/characters.md`) — `image_generate` is prompt-only, so the prompt text is the sole vehicle for character consistency
 3. Save to `prompts/NN-{cover|page}-[slug].md` using `write_file`
    - **Backup rule**: If prompt file exists, rename to `prompts/NN-{cover|page}-[slug]-backup-YYYYMMDD-HHMMSS.md`
 
@@ -257,8 +256,9 @@ Create image generation prompts for all pages.
 ## Visual Style
 Art: [art style] | Tone: [tone] | Layout: [layout type]
 
-## Character Reference
-[Character descriptions from characters/characters.md]
+## Character Reference (embedded inline — maintain exact traits below)
+- [Character A]: [detailed visual traits from characters/characters.md]
+- [Character B]: [detailed visual traits from characters/characters.md]
 
 ## Panel Breakdown
 [From storyboard.md - panel descriptions, actions, dialogue]
@@ -306,9 +306,9 @@ options:
 
 ## Step 7: Generate Images
 
-With confirmed prompts from Step 5/6, use the `image_generate` tool for all image rendering.
+With confirmed prompts from Step 5/6, use the `image_generate` tool. The tool accepts only `prompt` and `aspect_ratio` (`landscape` | `portrait` | `square`) and **returns a URL** — it does not accept reference images and does not write local files. Every invocation must be followed by a download step.
 
-**Aspect ratio mapping** — `image_generate` supports `landscape`, `portrait`, and `square`:
+**Aspect ratio mapping** — map the storyboard's `aspect_ratio` to the tool's enum:
 
 | Storyboard ratio | `image_generate` format |
 |------------------|-------------------------|
@@ -316,6 +316,12 @@ With confirmed prompts from Step 5/6, use the `image_generate` tool for all imag
 | `4:3`, `16:9`, `3:2` | `landscape` |
 | `1:1` | `square` |
 
+**Download procedure** (run after every successful `image_generate` call):
+
+1. Extract the `url` field from the tool result
+2. Fetch it to disk, e.g. `curl -fsSL "<url>" -o comic/{slug}/<target>.png`
+3. Verify the file is non-empty (`test -s <target>.png`); on failure, retry the generation once
+
 ### 7.1 Generate Character Reference Sheet (conditional)
 
 Character sheet is recommended for multi-page comics with recurring characters, but **NOT required** for all presets.
@@ -331,96 +337,56 @@ Character sheet is recommended for multi-page comics with recurring characters,
 **When generating**:
 1. Use Reference Sheet Prompt from `characters/characters.md`
 2. **Backup rule**: If `characters/characters.png` exists, rename to `characters/characters-backup-YYYYMMDD-HHMMSS.png`
-3. Call `image_generate` with `landscape` format → save to `characters/characters.png`
-4. **Compress** to reduce payload size when used as a reference:
-   - macOS: `sips -s format jpeg -s formatOptions 80 characters.png --out characters-compressed.jpg`
-   - Linux: `pngquant --quality=65-80 characters.png -o characters-compressed.png`
+3. Call `image_generate` with `landscape` format
+4. Download the returned URL → save to `characters/characters.png`
+
+**Important**: the downloaded sheet is for the **agent's own reference** when writing each page's prompt text below. `image_generate` cannot accept it as a visual input.
 
 ### 7.2 Generate Comic Pages
 
 **Before generating any page**:
 1. Confirm each prompt file exists at `prompts/NN-{cover|page}-[slug].md`
-2. Check whether `image_generate` accepts a reference image in the current runtime
-3. Determine if character sheet exists
-4. Choose the appropriate strategy below
+2. Confirm that each prompt has character descriptions embedded inline (see Step 5). `image_generate` is prompt-only, so the prompt text is the sole consistency mechanism.
 
-**Page Generation Strategy**:
+**Page Generation Strategy** (embed everything in the prompt text):
 
-| Character Sheet | `image_generate` reference support | Strategy |
-|-----------------|------------------------------------|----------|
-| Exists | Supported | **A**: Pass character sheet as reference with every page |
-| Exists | Not supported | **B**: Embed character descriptions in every prompt |
-| Skipped | — | **C**: Prompt file contains all descriptions inline |
+| Character sheet | Strategy |
+|-----------------|----------|
+| Exists | Use it as an agent-side reference when composing each prompt; embed the key traits inline in the prompt text |
+| Skipped | Prompt file already contains all descriptions inline |
 
-**Strategy A: Pass reference image**
-
-- For every page, read `prompts/NN-{type}-[slug].md` as the prompt input
-- Save output to `NN-{type}-[slug].png`
-- Use aspect ratio from storyboard (mapped to `landscape`/`portrait`/`square`)
-- Pass `characters/characters.png` (or compressed version) as the reference image
-
-**Reference failure recovery**:
-If generation fails when passing the reference:
-1. **Compress/convert** reference image:
-   - `sips -s format jpeg -s formatOptions 70 characters.png --out characters-compressed.jpg`
-   - Or reduce resolution: `sips -Z 1024 characters.png --out characters-small.png`
-2. **Retry** with compressed/converted image
-3. **If still fails**: Fall back to **Strategy C** — generate WITHOUT reference, with character descriptions embedded in prompt text
-
-**Strategy B: Embedding character descriptions in prompt**
-
-When reference images are not supported, create combined prompt files:
+**Example embedded prompt** (`prompts/01-page-xxx.md`):
 
 ```markdown
-# prompts/01-page-xxx.md (with embedded character reference)
+# Page 01: [Title]
 
-## Character Reference (maintain consistency)
-[Copy relevant sections from characters/characters.md here]
-- 大雄：Japanese boy, round glasses, yellow shirt, navy shorts...
-- 哆啦 A 梦：Round blue robot cat, white belly, red nose, golden bell...
+## Character Reference (embedded inline — maintain consistency)
+- 大雄：Japanese boy, round glasses, yellow shirt, navy shorts, worried expression...
+- 哆啦 A 梦：Round blue robot cat, white belly, red nose, golden bell, 4D pocket...
 
 ## Page Content
-[Original page prompt here]
+[Original page prompt body — panels, dialogue, visual metaphors]
 ```
 
-**Strategy C: Prompt-only (no character sheet)**
-
-When character sheet was skipped or the reference failed:
-- Prompt file already contains all character descriptions inline
-- No reference image needed
-- Rely on detailed text descriptions for character consistency
-
 **For each page (cover + pages)**:
 1. Read prompt from `prompts/NN-{cover|page}-[slug].md`
 2. **Backup rule**: If image file exists, rename to `NN-{cover|page}-[slug]-backup-YYYYMMDD-HHMMSS.png`
-3. Generate image via `image_generate` using Strategy A, B, or C
-4. Save to `NN-{cover|page}-[slug].png`
+3. Call `image_generate` with the prompt text and mapped aspect ratio
+4. Download the returned URL → save to `NN-{cover|page}-[slug].png`
 5. Report progress after each generation: "Generated X/N: [page title]"
 
 ---
 
-## Step 8: Merge to PDF
-
-After all images generated:
-
-```bash
-bun {baseDir}/scripts/merge-to-pdf.ts <comic-dir>
-```
-
-Where `{baseDir}` is this skill's directory. Creates `{topic-slug}.pdf` with all pages as full-page images.
-
----
-
-## Step 9: Completion Report
+## Step 8: Completion Report
 
 ```
 Comic Complete!
 Title: [title] | Art: [art] | Tone: [tone] | Pages: [count] | Aspect: [ratio] | Language: [lang]
 Location: [path]
+✓ source-{slug}.md (if content was pasted)
 ✓ analysis.md
 ✓ characters.png (if generated)
 ✓ 00-cover-[slug].png ... NN-page-[slug].png
-✓ {topic-slug}.pdf
 ```
 
 ---
@@ -429,9 +395,9 @@ Location: [path]
 
 | Action | Steps |
 |--------|-------|
-| **Edit** | Update prompt → Regenerate image → Regenerate PDF |
-| **Add** | Create prompt at position → Generate image → Renumber subsequent (NN+1) → Update storyboard → Regenerate PDF |
-| **Delete** | Remove files → Renumber subsequent (NN-1) → Update storyboard → Regenerate PDF |
+| **Edit** | Update prompt → Regenerate image → Download new PNG |
+| **Add** | Create prompt at position → Generate image → Download PNG → Renumber subsequent (NN+1) → Update storyboard |
+| **Delete** | Remove files → Renumber subsequent (NN-1) → Update storyboard |
 
 **File naming**: `NN-{cover|page}-[slug].png` (e.g., `03-page-enigma-machine.png`)
 - Slugs: kebab-case, unique, derived from content
diff --git a/skills/creative/baoyu-comic/scripts/merge-to-pdf.ts b/skills/creative/baoyu-comic/scripts/merge-to-pdf.ts
deleted file mode 100644
index bdd29c60c8..0000000000
--- a/skills/creative/baoyu-comic/scripts/merge-to-pdf.ts
+++ /dev/null
@@ -1,116 +0,0 @@
-import { existsSync, readdirSync, readFileSync } from "fs";
-import { join, basename } from "path";
-import { PDFDocument } from "pdf-lib";
-
-interface PageInfo {
-  filename: string;
-  path: string;
-  index: number;
-  promptPath?: string;
-}
-
-function parseArgs(): { dir: string; output?: string } {
-  const args = process.argv.slice(2);
-  let dir = "";
-  let output: string | undefined;
-
-  for (let i = 0; i < args.length; i++) {
-    if (args[i] === "--output" || args[i] === "-o") {
-      output = args[++i];
-    } else if (!args[i].startsWith("-")) {
-      dir = args[i];
-    }
-  }
-
-  if (!dir) {
-    console.error("Usage: bun merge-to-pdf.ts <comic-dir> [--output filename.pdf]");
-    process.exit(1);
-  }
-
-  return { dir, output };
-}
-
-function findComicPages(dir: string): PageInfo[] {
-  if (!existsSync(dir)) {
-    console.error(`Directory not found: ${dir}`);
-    process.exit(1);
-  }
-
-  const files = readdirSync(dir);
-  const pagePattern = /^(\d+)-(cover|page)(-[\w-]+)?\.(png|jpg|jpeg)$/i;
-  const promptsDir = join(dir, "prompts");
-  const hasPrompts = existsSync(promptsDir);
-
-  const pages: PageInfo[] = files
-    .filter((f) => pagePattern.test(f))
-    .map((f) => {
-      const match = f.match(pagePattern);
-      const baseName = f.replace(/\.(png|jpg|jpeg)$/i, "");
-      const promptPath = hasPrompts ? join(promptsDir, `${baseName}.md`) : undefined;
-
-      return {
-        filename: f,
-        path: join(dir, f),
-        index: parseInt(match![1], 10),
-        promptPath: promptPath && existsSync(promptPath) ? promptPath : undefined,
-      };
-    })
-    .sort((a, b) => a.index - b.index);
-
-  if (pages.length === 0) {
-    console.error(`No comic pages found in: ${dir}`);
-    console.error("Expected format: 00-cover-slug.png, 01-page-slug.png, etc.");
-    process.exit(1);
-  }
-
-  return pages;
-}
-
-async function createPdf(pages: PageInfo[], outputPath: string) {
-  const pdfDoc = await PDFDocument.create();
-  pdfDoc.setAuthor("baoyu-comic");
-  pdfDoc.setSubject("Generated Comic");
-
-  for (const page of pages) {
-    const imageData = readFileSync(page.path);
-    const ext = page.filename.toLowerCase();
-    const image = ext.endsWith(".png")
-      ? await pdfDoc.embedPng(imageData)
-      : await pdfDoc.embedJpg(imageData);
-
-    const { width, height } = image;
-    const pdfPage = pdfDoc.addPage([width, height]);
-
-    pdfPage.drawImage(image, {
-      x: 0,
-      y: 0,
-      width,
-      height,
-    });
-
-    console.log(`Added: ${page.filename}${page.promptPath ? " (prompt available)" : ""}`);
-  }
-
-  const pdfBytes = await pdfDoc.save();
-  await Bun.write(outputPath, pdfBytes);
-
-  console.log(`\nCreated: ${outputPath}`);
-  console.log(`Total pages: ${pages.length}`);
-}
-
-async function main() {
-  const { dir, output } = parseArgs();
-  const pages = findComicPages(dir);
-
-  const dirName = basename(dir) === "comic" ? basename(join(dir, "..")) : basename(dir);
-  const outputPath = output || join(dir, `${dirName}.pdf`);
-
-  console.log(`Found ${pages.length} pages in: ${dir}\n`);
-
-  await createPdf(pages, outputPath);
-}
-
-main().catch((err) => {
-  console.error("Error:", err.message);
-  process.exit(1);
-});