Baoyu Comic Skill for Hermes Agent: Create Knowledge Comics from Any Content

The Baoyu Comic skill turns text, URLs, or topics into illustrated knowledge comics. You pick an art style and tone, the agent handles storyboarding, prompt generation, and image rendering. This is for anyone who wants to explain a concept, tell a biography, or make a tutorial as a comic strip, without touching an image editor.

What it does

You give the agent source material (a Wikipedia article, a PDF, a URL, or just a topic) and optionally an art style, tone, layout, and language. The skill produces a full comic: a storyboard, character definitions, per-page image prompts, and rendered PNG files. It supports partial workflows so you can stop after the storyboard, after the prompts, or regenerate specific pages.

Before you start

• The skill is optional and must be installed on demand. Its path is optional-skills/creative/baoyu-comic.
• It runs on Linux, macOS, and Windows.
• The agent must have the image_generate tool available. That tool is prompt-only: it accepts a text prompt and an aspect ratio, and returns a URL. It does not accept reference images.
• The agent needs curl or equivalent to download generated images from URLs to local files.
• The user should have a clear topic or source content ready. The skill can accept text, file paths, URLs, or just a topic name.

Workflow overview

The skill follows a linear pipeline with two optional review gates. The agent tracks progress with a checklist.

Comic Progress:
- [ ] Step 1: Setup & Analyze
  - [ ] 1.1 Analyze content
  - [ ] 1.2 Check existing directory
- [ ] Step 2: Confirmation - Style & options ⚠️ REQUIRED
- [ ] Step 3: Generate storyboard + characters
- [ ] Step 4: Review outline (conditional)
- [ ] Step 5: Generate prompts
- [ ] 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 descriptions embedded in prompt)
- [ ] Step 8: Completion report

Input → Analyze → [Check Existing?] → [Confirm: Style + Reviews] → Storyboard → [Review?] → Prompts → [Review?] → Images → Complete

Step 1: Setup and analyze

The agent reads the source content, saves it as source-{slug}.md, and writes a content analysis to analysis.md. The slug is 2-4 kebab-case words from the topic, for example alan-turing-bio. If a directory with that slug already exists, the agent appends a timestamp to avoid overwriting.

Step 2: Confirm style, focus, audience, and review preferences

This step is required and cannot be skipped. The agent uses the clarify tool to ask the user one question at a time. The questions cover art style, tone, layout, aspect ratio, language, whether to review the storyboard, and whether to review the prompts. Because clarify handles one question per call, the agent asks the most important question first and proceeds sequentially.

Timeout handling is critical. If clarify returns a timeout message, the agent treats that as a default for that one question only. It continues asking the remaining questions. The agent must surface the default visibly in the next message so the user can correct it later. For example: "Style: defaulted to ohmsha preset (clarify timed out). Say the word to switch." The agent must not collapse all remaining questions into a single "use all defaults" pass.

Step 3: Generate storyboard and characters

The agent writes a storyboard to storyboard.md and character definitions to characters/characters.md. The storyboard breaks the content into pages and panels. Character definitions describe appearance, personality, and role in text.

Step 4: Review outline (conditional)

Only if the user requested a review in Step 2. The agent presents the storyboard and character definitions for approval before proceeding.

Step 5: Generate prompts

The agent writes one prompt file per page under prompts/. Each file is named NN-{cover|page}-[slug].md. The prompt includes the scene description, character descriptions from characters/characters.md, and the chosen art style, tone, and layout. This is the reproducibility record.

Step 6: Review prompts (conditional)

Only if the user requested a review in Step 2. The agent presents the prompts for approval before generating images.

Step 7: Generate images

This step has two sub-steps.

7.1 Generate character sheet

The agent generates a character reference sheet as characters/characters.png using image_generate with aspect ratio landscape. This is recommended for multi-page comics with recurring characters. It can be skipped for simple presets like four-panel minimalist or single-page comics. The prompt file at characters/characters.md must exist before calling image_generate. The rendered PNG is a human-facing review artifact. It does not drive Step 7.2 because image_generate cannot accept images as visual input. Page prompts already embed character descriptions from characters/characters.md.

7.2 Generate pages

For each page, the agent calls image_generate with the prompt from the corresponding file in prompts/ and the aspect ratio mapped from the storyboard. The mapping is:

Storyboard ratio	image_generate format
3:4, 9:16, 2:3	portrait
4:3, 16:9, 3:2	landscape
1:1	square

After each call, the agent reads the URL from the tool result, fetches the image bytes using an absolute output path, and verifies the file exists and is non-empty before proceeding.

Never rely on shell CWD persistence for -o paths. The terminal tool's persistent-shell CWD can change between batches. curl -o relative/path.png is a silent footgun. Always pass a fully-qualified absolute path to -o, or pass workdir= to the terminal tool. An incident in April 2026 caused pages 06-09 of a 10-page comic to land at the repo root instead of the comic directory because batch 3 inherited a stale CWD.

Step 8: Completion report

The agent presents a summary of what was created, including file paths and counts.

Options

Visual Dimensions

Option	Values	Description
Art	ligne-claire (default), manga, realistic, ink-brush, chalk, minimalist	Art style / rendering technique
Tone	neutral (default), warm, dramatic, romantic, energetic, vintage, action	Mood / atmosphere
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 used for style / palette trait extraction (not passed to the image model). See Reference Images above.

Partial Workflow Options

Option	Description
Storyboard only	Generate storyboard only, skip prompts and images
Prompts only	Generate storyboard + prompts, skip images
Images only	Generate images from existing prompts directory
Regenerate N	Regenerate specific page(s) only (e.g., 3 or 2,5,8)

Details: references/partial-workflows.md

Art, Tone & Preset Catalogue

• Art styles (6): ligne-claire, manga, realistic, ink-brush, chalk, minimalist. Full definitions at references/art-styles/.md.

• Tones (7): neutral, warm, dramatic, romantic, energetic, vintage, action. Full definitions at references/tones/.md.

• Presets (5) with special rules beyond plain art+tone:

  Preset	Equivalent	Hook
  ohmsha	manga + neutral	Visual metaphors, no talking heads, gadget reveals
  wuxia	ink-brush + action	Qi effects, combat visuals, atmospheric
  shoujo	manga + romantic	Decorative elements, eye details, romantic beats
  concept-story	manga + warm	Visual symbol system, growth arc, dialogue+action balance
  four-panel	minimalist + neutral + four-panel layout	起承转合 structure, B&W + spot color, stick-figure characters

  Full rules at references/presets/.md, load the file when a preset is picked.

• Compatibility matrix and content-signal → preset table live in references/auto-selection.md. Read it before recommending combinations in Step 2.

File Structure

Output directory: comic/{topic-slug}/

• Slug: 2-4 words kebab-case from topic (e.g., alan-turing-bio)
• Conflict: append timestamp (e.g., turing-story-20260118-143052)

Contents:

File	Description
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 (downloaded from image_generate)
prompts/NN-{cover|page}-[slug].md	Generation prompts
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

Detection Priority:

1. User-specified language (explicit option)
2. User's conversation language
3. Source content language

Rule: Use user's input language for ALL interactions:

• Storyboard outlines and scene descriptions
• Image generation prompts
• User selection options and confirmations
• Progress updates, questions, errors, summaries

Technical terms remain in English.

Reference Images

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 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

Usage modes (per reference):

Usage	Effect
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:

references:
  - ref_id: 01
    filename: 01-ref-scene.png
    usage: style
    traits: "muted earth tones, soft-edged ink wash, low-contrast backgrounds"

Character consistency is driven by text descriptions in characters/characters.md (written in Step 3) that get embedded inline in every page prompt (Step 5). The optional PNG character sheet generated in Step 7.1 is a human-facing review artifact, not an input to image_generate.

Page Modification

Action	Steps
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 absolute paths for curl -o, never rely on persistent-shell CWD across batches. Silent footgun: files land in the wrong directory and subsequent ls on the intended path shows nothing. See Step 7 "Download step".
• 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. The PNG is a review/regeneration aid; page prompts (written in Step 5) use the text descriptions in characters/characters.md, not the PNG. image_generate does not accept images as visual input
• Strip secrets, scan source content for API keys, tokens, or credentials before writing any output file

When not to use it

This skill is designed for educational, biographical, and tutorial content. It is not suited for purely fictional narratives, abstract concepts without a clear linear structure, or content that requires photorealistic rendering. The image_generate tool is prompt-only and cannot use reference images as visual input, so it cannot replicate a specific character design from an uploaded image.

Limits and gotchas

• The image_generate tool returns a URL, not a file. You must download every image to the output directory.
• The agent must use absolute paths for curl -o because the persistent-shell CWD can change between batches.
• The character sheet PNG is a review artifact only. It does not influence image generation because image_generate cannot accept images.
• Step 2 confirmation is required. The agent cannot skip it even if the user seems to want defaults.
• Timeouts on clarify questions must be handled one at a time, not collapsed.
• Secrets in source content must be stripped before writing any output file.

What pairs with this

The skill references several templates and guides that live in the same repository:

• analysis-framework.md - Deep content analysis
• character-template.md - Character definition format
• storyboard-template.md - Storyboard structure
• ohmsha-guide.md - Ohmsha manga specifics
• workflow.md - Full workflow details
• auto-selection.md - Content signal analysis
• partial-workflows.md - Partial workflow options

These files define the art styles, tones, presets, and layouts referenced in the options table. The agent should load the relevant file when a preset is picked.