Agent skill

Recurring Character Diary Comic

by ZSeven-W in ZSeven-W/craft-skills

Create, audit, and repair short page-native diary comics around an existing authorized recurring character, with story-directed page rhythm, exact dialogue, directional-surface proof, and…

Apache-2.0Auto-check passedMedia & Creative

Install Recurring Character Diary Comic

skills CLI
$ npx skills add ZSeven-W/craft-skills --skill recurring-character-diary-comic -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install ZSeven-W/craft-skills recurring-character-diary-comic --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/ZSeven-W/craft-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/recurring-character-diary-comic .claude/skills/recurring-character-diary-comic && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
recurring-character-diary-comic
GitHub stars
225
Used in
1 other repo
Token cost
~2.8k tokens
SKILL.md length
1,406 words
Files
17 (incl. scripts, references)
Skills in repo
5
Repo updated
First seen
Licence
Apache-2.0

At a glance

Create, audit, and repair short page-native diary comics around an existing authorized recurring character, with story-directed page rhythm, exact dialogue, directional-surface proof, and…

  • Works in 5 steps: Use only original characters,… → Refuse requests to imitate a named… → Normalize the recurring identity into a… → …
  • Extending an established recurring-character series from an anecdote
  • SKILL.md covers Load only the needed references, Choose the mode, Establish rights and identity and Lock the story and page plan, plus 5 more sections
  • Runs Python scripts from its folder

What it does

Recurring Character Diary Comic is an agent skill from ZSeven-W/craft-skills. Create, audit, and repair short page-native diary comics around an existing authorized recurring character, with story-directed page rhythm, exact dialogue, directional-surface proof, and original-resolution visual QA. Use when extending an established recurring-character series from an anecdote, conversation, dream, or viewpoint, or when checking or repairing identity, anatomy, text, prop, relation, continuity, or layout defects. Do not use for standalone character design, educational explainers, infographics…

Its SKILL.md is about 2.8k tokens, which your agent loads only when the skill is triggered. The skill folder holds 20 other files, including scripts and reference files (for example `README.en.md`, `README.md` and `agents/openai.yaml`).

It sits in Media & Creative, covering Comics and storyboards, Journaling and reflection and Visual regression testing. The repository describes itself as: Research-backed, eval-driven skills for AI agents. The licence is Apache-2.0.

When your agent uses it

  • Extending an established recurring-character series from an anecdote
  • Repairing identity
  • Standalone character design
  • Educational explainers

Example prompts

  • “/recurring-character-diary-comic”

Requirements

  • Python 3

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Use only original characters, public-domain material, or references the user confirms they may use.
  2. Refuse requests to imitate a named living artist, use a copyrighted franchise character without authorization, or reproduce a third-party…
  3. Normalize the recurring identity into a profile before page generation. Separate stable identity invariants, episode variables, exact…
  4. Treat a standalone reference for a required mark or accessory as more authoritative than an incidental depiction inside a character sheet.
  5. If no usable recurring identity exists, stop and return the profile requirements. Do not invent a permanent character while producing the…

What it can do on your machine

Read from SKILL.md and the folder at commit 01c8ffe. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 3 files in scripts/ (Python), which the agent can run.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Recurring Character Diary Comic loads about 2.8k tokens when it runs, and up to ~20k if it reads all its reference files. Until then it costs about 175 tokens; SKILL.md has 1,406 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~175
When it runs · the whole SKILL.md, loaded when a task matches
~2.8k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~20k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from ZSeven-W/craft-skills at commit 01c8ffe, republished under its Apache-2.0 licence (© ZSeven-W). 1,406 words, ~2,793 tokens.

Download SKILL.mdSave it as .claude/skills/recurring-character-diary-comic/SKILL.md (or your agent's skills folder). This skill also uses 16 other files; get the full folder from GitHub.
name
recurring-character-diary-comic
description
Create, audit, and repair short page-native diary comics around an existing authorized recurring character, with story-directed page rhythm, exact dialogue, directional-surface proof, and original-resolution visual QA. Use when extending an established recurring-character series from an anecdote, conversation, dream, or viewpoint, or when checking or repairing identity, anatomy, text, prop, relation, continuity, or layout defects. Do not use for standalone character design, educational explainers, infographics, one-off illustrations, generic memes, unauthorized characters or marks, imitation of a named living artist, watermarking, or social-platform publishing.

Recurring Character Diary Comic

Turn a concrete observation into one coherent comic page while preserving an authorized recurring identity. The default is page-native generation: the image model solves the complete page, not a collection of independent cards.

Load only the needed references

Choose the mode

  • Create: require an existing authorized recurring identity, pass the premise gate, lock story and visual evidence, compare three page structures, generate one complete page, and inspect the final pixels.
  • Audit: inspect an existing page against its approved character and story records; report evidence without silently editing it.
  • Repair: audit first, name the failing field, and prefer one bounded local page edit. Reconstruct panels only under the fallback rule below.

Do not expand the request into character design, branding, publishing, or an unrelated illustration task.

Establish rights and identity

  1. Use only original characters, public-domain material, or references the user confirms they may use.
  2. Refuse requests to imitate a named living artist, use a copyrighted franchise character without authorization, or reproduce a third-party signature mark. Offer medium-level traits instead.
  3. Normalize the recurring identity into a profile before page generation. Separate stable identity invariants, episode variables, exact optional marks, and forbidden drift.
  4. Treat a standalone reference for a required mark or accessory as more authoritative than an incidental depiction inside a character sheet.
  5. If no usable recurring identity exists, stop and return the profile requirements. Do not invent a permanent character while producing the first episode.

Lock the story and page plan

  1. Apply the premise gate in references/story-rhythm.md. Prefer a drawable behavior over an abstract slogan, advertisement, or moral.
  2. Preserve the user's experience and viewpoint. Extract setup, visible progression, reaction, and the final observation or reveal.
  3. Choose 4–8 beats because the story needs them, not to fill a grid. Give supporting characters, objects, screens, and environments narrative jobs.
  4. Lock every approved string before generation: exact spelling, punctuation, speaker, bubble ownership, and reading order.
  5. Give each character, prop, state, panel, and critical relation a stable semantic ID. Classify requirements as blocking story truth (S0), blocking identity truth (S1), or replaceable composition preference (S2).
  6. Express critical relations as visible evidence rather than co-occurrence. Record subject, predicate, object, contact or direction, visible proof, forbidden ambiguity, and required next state.
  7. For every screen, paper, receipt, ticket, book, sign, or control, record its reader/operator, content face, reader side, viewer side, hinge or controls, gaze/hands, and the pixels that prove the relation. A nearby character and object do not prove reading or operation.
  8. Draw three low-fidelity page-skeleton candidates before art generation. They must differ structurally, not merely by small frame offsets. Show beat IDs, reading path, relative area, visual anchor, gutters, negative space, and any content-derived border interaction.
  9. Compare all three at working size and exactly 25%. Select the structure that preserves the story evidence and creates the clearest rhythm. Do not default to an equal card grid, a repeated row stack, or random rotation.
  10. Border breaks, foreground overflow, semantic insets, open frames, and shared boundaries are optional vocabulary, never quotas. Use an effect only when its subject and narrative reason are clear and it does not hide faces, hands, text, direction, or state evidence.

Choose the production route

Risk changes prompt and QA attention; it does not automatically decompose the page.

  • page-native — default. Generate the entire page in one image, including panel topology, scenery, character scale, props, border interactions, bubbles, exact dialogue, paper texture, palette, and negative space.
  • page-native-unlettered — text fallback. Use only when the complete page direction passes but rendered dialogue cannot be made exact. Generate one complete bubble-free, text-free page with intentional dialogue space, then add deterministic bubbles and text without cutting the art into panel sources.
  • panel-reconstruction — disclosed fallback. Use only after two page-native attempts fail the same named S0 or S1 field, one local edit cannot isolate the defect, and the user explicitly accepts the loss of whole-page coherence. Generate and freeze independent panel sources, then reconstruct and re-audit the page. Do not switch merely because independent panels are easier to validate.

If reconstruction would destroy the requested page language, stop and report the blocker instead of delivering a compliant collage.

Show full SKILL.md (615 more words)Show less

Generate the page

  1. Label each image input by role: identity, exact mark, medium, edit target, or supporting reference. Do not use a reference with a known defect as authority for the same field.
  2. Compile one concise whole-page prompt containing the selected skeleton, beats in reading order, identity invariants, state continuity, exact dialogue and speakers, medium, and only ambiguities that can break S0 or S1.
  3. Use the same accepted identity and medium references throughout the run. Replacing them starts a successor run and requires rechecking affected artifacts.
  4. Use no more than two page-native art attempts before a route decision. If the first result collapses into an unrelated row stack or card grid, correct the structural reference or prompt; do not disguise it with cosmetic diagonals.
  5. Inspect every candidate at original resolution and exactly 25%. A generated file or successful tool call is not an acceptance result.
  6. When one named defect can be isolated, prefer one constrained local edit and compare all pixels outside the allowed region against the saved baseline.
  7. If only text fails, use page-native-unlettered. Preserve the whole-canvas art as one borderless source and render only approved bubbles and text.
  8. A panel-reconstruction fallback receives a new artifact stage and new two-axis review. It never inherits a page-native artistic pass.

Handle dialogue reliably

  1. A page-native final may contain model-rendered dialogue only when every approved character, speaker, order, contour, and protected region passes on the final pixels.
  2. For deterministic lettering, use one complete unlettered page as the full-canvas source. Do not slice it into narrative panels.
  3. Speech bubbles must be connected, closed silhouettes with one outline. Prefer short, soft-rounded tails; reject detached tails, needle tips, narrow black necks, internal body arcs, seams, double outlines, or open contours.
  4. Compare rendered strings character by character with the locked script. Reject rewritten wording, extra text, wrong speakers, ambiguous order, or action-obscuring bubbles.
  5. Record art and lettering as separate artifact stages with input, font, geometry, and output hashes when deterministic rendering is used.

Inspect and repair

  1. Declare the artifact stage before judging it: identity-reference, page-native-final, page-native-unlettered, unlettered-panel, reconstructed-page, lettered-final, or repair-candidate.
  2. Enumerate every panel's cast, visible hands, story-critical props, state changes, directional surfaces, physical relations, and approved strings.
  3. Verify positive visible evidence for each relation. Trace contact geometry, current reader/operator, content face, gaze/hands, direction, and required next state.
  4. Run contract-fidelity QA on the exact final hash at original resolution. Record pass, fail, or not-verified; every S0 or S1 failure blocks acceptance.
  5. Separately run editorial-layout QA on the same hash at original size and 25%. Check reading path, beat hierarchy, panel rhythm, border language, intentional negative space, final emphasis, page-wide line/palette coherence, and accidental card-grid signals.
  6. A page is showcase-ready only when both axes pass. Technical acceptance, editorial acceptance, user approval, and publication authority remain separate states.
  7. After repair, recheck the repaired region and the complete final page. Reject a local edit that changes pixels outside its allowed region.
  8. Stop on explicit acceptance, the bounded attempt limit, or two consecutive attempts that fail to improve the named field without damaging locked content. Report the blocker rather than weakening the contract.

Deliver evidence

For Create, deliver the character/profile identifier, locked story and visual contract, three page-skeleton candidates and selected strategy, production route and attempt ledger, accepted final hash, any unlettered/repair/fallback artifacts actually used, and the two-axis QA record. Do not manufacture panel-source deliverables for a page-native run.

For Audit or Repair, deliver the exact inspected or repaired artifact hash, evidence locators, state, and next safe action.

Keep states distinct: planned, generated, inspected, accepted, showcase-ready, internal-only, and user-approved. None grants permission to publish.

© ZSeven-W, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 16 other files (scripts, references) in skills/recurring-character-diary-comic of ZSeven-W/craft-skills.

  • SKILL.md
  • README.en.md
  • README.md
  • agents/openai.yaml
  • references/character-profile-schema.md
  • references/compositor-manifest.md
  • references/continuity-and-visual-qa.md
  • references/editorial-layout-gate.md
  • references/story-rhythm.md
  • references/visual-task-contract.md
  • scripts/compose_panels.py
  • scripts/requirements.txt
  • scripts/self_test_compositor.py
  • templates/compositor-manifest.example.json
  • templates/editorial-layout-gate.example.yaml
  • templates/page-native-lettering-manifest.example.json
  • templates/visual-task-contract.example.yaml

Open the folder on GitHubat commit 01c8ffe

Used in 1 other repository

We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in ZSeven-W/craft-skills, which our catalogue first saw on October 7, 2026.

Compare with similar skills

Recurring Character Diary Comic next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Recurring Character Diary Comic compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Recurring Character Diary Comic this skillZSeven-W/craft-skills2251 repos~2.8kAutomated safety check: PassApache-2.0
Direct Image Creationopenvetta/open-vetta290—~1.2kAutomated safety check: PassApache-2.0
Anygen StorybookLeoYeAI/openclaw-master-skills2.2k—~4.3kAutomated safety check: PassMIT
Canghe Comicfreestylefly/canghe-skills4618 repos~3.2kAutomated safety check: PassNone
Visual Explainernicobailon/visual-explainer10k—~3.5kAutomated safety check: PassMIT
Lanshu Create AI Presenter Videocclank/lanshu-create-ai-presenter-video2.6k—~3.6kAutomated safety check: PassMIT

Similar skills

  • Direct Image Creation

    openvetta/open-vetta

    Turn a creative request into a production-ready AI image brief, reference plan, node workflow, and model-profile prompt.

    290 GitHub stars~1.2k tokensUpdated 2 days ago
    Media & CreativeAuto-check passed
  • Anygen Storybook

    LeoYeAI/openclaw-master-skills

    Use this skill any time the user wants to create visual stories, illustrated narratives, or storybook content.

    2.2k GitHub stars~4.3k tokensUpdated 2 mo ago
    Frontend & DesignAuto-check passed
  • Canghe Comic

    freestylefly/canghe-skills

    Knowledge comic creator supporting multiple art styles and tones.

    461 GitHub starsUsed in 8 repos~3.2k tokens
    Media & CreativeAuto-check passed
  • Visual Explainer

    nicobailon/visual-explainer

    Turns systems, code changes, plans and data into self-contained HTML pages where figures carry the explanation, with options for slide decks and narrated MP4 videos.

    10k GitHub stars~3.5k tokensUpdated today
    Media & CreativeAuto-check passed
  • Lanshu Create AI Presenter Video

    cclank/lanshu-create-ai-presenter-video

    Turn a topic or finished script into a complete, publish-ready explainer video — led by an AI presenter from an authorized adult presenter image, or performed in one of nine visual explainer styles…

    2.6k GitHub stars~3.6k tokensUpdated today
    Media & CreativeAuto-check passed
  • Tutti UI System

    tutti-os/tutti

    A skill your agent uses when working with @tutti-os/ui-system components, replacing local UI with shared components, querying component ids or metadata, promoting UI into shared base or business…

    3.8k GitHub stars~2.7k tokensUpdated 3 days ago
    Media & CreativeAuto-check passed

More from ZSeven-W/craft-skills

  • Native Transparent Imagegen

    ZSeven-W/craft-skills

    Generate new raster assets that must contain native pixel transparency, then verify the untouched PNG or WebP before delivery.

    225 GitHub starsUsed in 1 repo~1.1k tokens
    Auto-check passed
  • Qwen Image Gen

    ZSeven-W/craft-skills

    为 Qwen-Image 编写和改写生图提示词,区分局部编辑与参考主体创作,处理同人物系列、画幅及原图验收;用户要求生成时连接其已有的 Qwen 图像工作流。不用于下载权重、通用模型运维或其他图像模型。

    225 GitHub stars~478 tokensUpdated 18 days ago
    Auto-check passed
  • Logo Semantic Fusion

    ZSeven-W/craft-skills

    Develop and evaluate logo concepts when the explicit design problem is semantic fusion: two or more brand meanings must share a contour, stroke, negative space, glyph skeleton, or shape system.

    225 GitHub starsUsed in 1 repo~1.7k tokens
    Auto-check passed
  • Single Path Process Diorama

    ZSeven-W/craft-skills

    将 3–5 步线性流程制作成单张完整、非宫格的微缩剧场成品。适用于流程故事封面、制作过程图、软件交付旅程与科普式微缩场景;以单一路径、可追踪载体、可辨识动作和题材匹配的材质表达顺序。不用于需要精确分支和数据尺度的流程图、实际操作规程或普通城市微缩景观。

    225 GitHub stars~824 tokensUpdated 18 days ago
    Auto-check passed

Questions about Recurring Character Diary Comic

What does Recurring Character Diary Comic do?

Create, audit, and repair short page-native diary comics around an existing authorized recurring character, with story-directed page rhythm, exact dialogue, directional-surface proof, and…. Recurring Character Diary Comic is an agent skill from ZSeven-W/craft-skills. Create, audit, and repair short page-native diary comics around an existing authorized recurring character, with story-directed page rhythm, exact dialogue, directional-surface proof, and original-resolution visual QA.

When should I use Recurring Character Diary Comic?

Recurring Character Diary Comic fits situations like: extending an established recurring-character series from an anecdote; repairing identity; standalone character design; educational explainers.

How do I install Recurring Character Diary Comic in Claude Code?

Run `npx skills add ZSeven-W/craft-skills --skill recurring-character-diary-comic -a claude-code`. Or copy the skill folder (skills/recurring-character-diary-comic in ZSeven-W/craft-skills) into .claude/skills/recurring-character-diary-comic in your project. Claude Code loads it when a task matches its description.

How do I install Recurring Character Diary Comic in Codex?

Run `npx skills add ZSeven-W/craft-skills --skill recurring-character-diary-comic -a codex`. Or copy the skill folder (skills/recurring-character-diary-comic in ZSeven-W/craft-skills) into .agents/skills/recurring-character-diary-comic in your project. Codex loads it when a task matches its description.

Can I use Recurring Character Diary Comic in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ZSeven-W/craft-skills --skill recurring-character-diary-comic -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/recurring-character-diary-comic, .gemini/skills/recurring-character-diary-comic, .github/skills/recurring-character-diary-comic and .opencode/skills/recurring-character-diary-comic in your project.

What does Recurring Character Diary Comic need to run?

Going by SKILL.md and its folder, Recurring Character Diary Comic needs Python for the scripts in its folder. Our summary lists: Python 3.

Does Recurring Character Diary Comic access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Recurring Character Diary Comic safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Recurring Character Diary Comic use?

Recurring Character Diary Comic is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Recurring Character Diary Comic use?

About 2.8k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 18k tokens, read only when the agent opens those files.

What are the alternatives to Recurring Character Diary Comic?

Skills that share tags, products or a category with Recurring Character Diary Comic: Direct Image Creation (openvetta/open-vetta, 290 stars), Anygen Storybook (LeoYeAI/openclaw-master-skills, 2.2k stars), Canghe Comic (freestylefly/canghe-skills, 461 stars) and Visual Explainer (nicobailon/visual-explainer, 10k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Recurring Character Diary Comic?

ZSeven-W (a GitHub organization) maintains it in ZSeven-W/craft-skills, which has 225 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on September 22, 2026.

Source: ZSeven-W/craft-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.