Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/wenn-id/comicsol/comic-solnpx skills add wenn-id/comicsol --skill comic-solgit clone --depth 1 https://github.com/wenn-id/comicsolWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/wenn-id/comicsol/comic-sol)<a href="https://agentmods.dev/skills/wenn-id/comicsol/comic-sol"><img src="https://agentmods.dev/badge/skills/wenn-id/comicsol/comic-sol.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00075 | $0.03022 |
| Opus 5 | $0.00037 | $0.01511 |
| Sonnet 5 | $0.00015 | $0.00604 |
| Haiku 4.5 | $0.00007 | $0.00302 |
Grade A, and why
comic-sol scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
100% identical to comic-sol — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 220 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Comic Sol
Turn one natural-language request into a local, editable comic project. Reason about story and images, then delegate deterministic work to the bundled Python scripts.
Read progressively
- Read workflow for input detection, all ten stages, commands, state transitions, failures, resume, and completion.
- Read creative direction before authoring plans, character fingerprints, storyboards, references, or image prompts.
- Read capability detection before the first
doctorcall and again only if the exposed tools change. - Read image provider setup before agent-managed external image tools; its ComfyUI route is reference/experimental and outside the deterministic engine.
- Read visual QA before accepting, retrying, overriding, composing, or exporting any generated panel.
- Read safety and IP before sending prompts externally and whenever people, minors, sensitive data, named styles, franchises, or refusals appear.
- Read schemas whenever writing or revising JSON artifacts.
Core orchestration
- Detect resume/source-file/pasted-story/short-prompt mode in the normative order.
- Ask only a materially required workflow question; otherwise apply defaults.
- Inspect exposed tool metadata without invoking tools. Pass the best usable image tool as
agent-image-generationwith its declared features todoctor; passunavailablefor an inspectable inventory with no usable tool, and no flags if inspection is unavailable or fails. Never infer capability availability or features from provider, model, or tool names. - Run doctor, initialize or inspect the project, then validate and record every stage.
Capability warnings allow deterministic planning. If neither a declared native tool nor
a declared external adapter is available, prepare that handoff and then transition
the project to
BLOCKEDuntil a destination declares a usable capability. Deterministic scripts never discover or call providers. - Generate canonical references and panels into attempt paths, including through an explicitly selected agent-managed external adapter; normal intake, retention, review, and promotion gates still apply. For prepared-handoff execution, run
PYTHON scripts/comic_sol.py handoff inspect PROJECT(installed equivalent:comic-sol handoff inspect PROJECT), select only a reported job whose effectivestatusisready, and passPROJECT/<jobs[].path>using the exactpathreturned for that job. After result intake, inspect again before retrying; the new inspection is authoritative for status and the next attempt. Never enumerate or execute retainedgeneration/jobs/*.jsonfiles directly. See workflow for the full lifecycle. Require the image model to draw each exactgenerated-visualstoryboard SFX and never adeterministic-letteringone. Inspect every result visually, record all seven QA checks, and repair only failures within budget. Route a bad effect to lettering withsfx_repair.pyand follow itsnext_action: ink the model drew needs aregeneratereview, not re-lettering alone. - Promote accepted attempts; deterministically letter dialogue, captions, and any
deterministic-letteringSFX while validating and counting exactgenerated-visualstoryboard SFX without drawing it in Pillow; compose pages, inspect every composed page and write itsqa/pages/page-{NNN}.jsonrecord, export the PDF, render the QA report (which projects the terminal status), validate final integrity, then transition to it. - Return status, counts, warnings, and clickable project output paths.
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- assets/fonts/ComicNeue-Bold.ttf 54 KB
- assets/fonts/ComicNeue-Regular.ttf 56 KB
- assets/fonts/NotoSans-Regular.ttf 556 KB
- assets/fonts/OFL-ComicNeue.txt 4.3 KB
- assets/fonts/OFL-NotoSans.txt 4.3 KB
- references/capability-detection.md 4.3 KB
- references/creative-direction.md 3.0 KB
- references/image-provider-setup.md 10 KB
- references/safety-ip.md 2.1 KB
- references/schemas.md 79 KB
- references/starter-templates.md 3.4 KB
- references/visual-qa.md 11 KB
- references/workflow.md 23 KB
- scripts/__init__.py 56 B runs code
- scripts/character_identity.py 31 KB runs code
- scripts/character_quality.py 31 KB runs code
- scripts/comic_sol.py 210 KB runs code
- scripts/command_service.py 11 KB runs code
- scripts/compose_pages.py 15 KB runs code
- scripts/core_primitives.py 14 KB runs code
- scripts/dogfood_report.py 31 KB runs code
- scripts/export_pdf.py 12 KB runs code
- scripts/font_cmap.py 5.8 KB runs code
- scripts/font_coverage.py 28 KB runs code
- scripts/handoff_archive.py 45 KB runs code
- scripts/handoff.py 41 KB runs code
- scripts/input_limits.py 7.7 KB runs code
- scripts/layouts.py 5.0 KB runs code
- scripts/letter_panels.py 54 KB runs code
- scripts/lifecycle_contracts.py 1.5 KB runs code
- scripts/normalize_panels.py 8.7 KB runs code
- scripts/page_quality.py 65 KB runs code
- scripts/pdf_quality.py 7.6 KB runs code
- scripts/project_io.py 50 KB runs code
- scripts/quality_records.py 2.5 KB runs code
- scripts/quality_sample.py 4.7 KB runs code
- scripts/raster_limits.py 217 B runs code
- scripts/reference_strategy.py 25 KB runs code
- scripts/render_report.py 25 KB runs code
- scripts/repair_strategy.py 42 KB runs code
- scripts/schema.py 8.5 KB runs code
- scripts/sfx_repair.py 16 KB runs code
- scripts/sfx_verification.py 19 KB runs code
- scripts/stage_registry.py 3.2 KB runs code
- scripts/starter_templates.py 5.7 KB runs code
- scripts/typography.py 18 KB runs code
- scripts/validate_project.py 144 KB runs code
- scripts/version.py 95 B runs code
- templates/character-bible.json 50 B
- templates/character-identity-pack.json 50 B
- templates/manifest.json 1.2 KB
- templates/page-qa.json 2.8 KB
- templates/panel-record.json 2.4 KB
- templates/qa-report.md.tmpl 464 B
- templates/starters/v1/action-focused/plan/character-bible.json 1.1 KB
- templates/starters/v1/action-focused/plan/story-plan.json 1.3 KB
- templates/starters/v1/action-focused/plan/storyboard.json 7.6 KB
- templates/starters/v1/action-focused/source/input.txt 171 B
- templates/starters/v1/action-focused/source/request.json 49 B
- templates/starters/v1/dialogue-two-page/plan/character-bible.json 2.0 KB
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 220 lines · 75 tokens per session scan A 3f1a0254d26a
comic-sol is a skill published in the GitHub repository wenn-id/comicsol (5 stars, last pushed 4d ago), licensed MIT. It adds 75 tokens to every session and 3,022 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to comic-sol, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
hanxiaoyu-visual-story-comic
Turn original, public-domain, or user-authorized stories, scripts, articles, and outlines into coherent multi-page comics with stable characters, scenes, props, page plans, image prompts, optional image generation, and continuity review. Use when the user asks for 漫画、条漫、绘本式故事页、连续分镜图、story comic、webtoon, comic pages…
comic-strip
Write a comic strip in the Comic Chat studio format and return a link that opens it, rendered, in the editor. Use when the user wants a comic, a cartoon, or illustrated panels for a blog post, a slide, a changelog, a README, or a recap.
general-video
Author or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated…
media-ingest
Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.
diagnostic-stem-delivery
Audio production with diagnostic analysis, timecode parsing from documents, and verified export workflow.
chengfeng-cut
剪辑中文口播原素材:逐词转录、词典修字出修字表、五轮扫描找口误与重复、汇总表与重复句子表、打开 Studio 让用户复核、复盘沉淀用户偏好与词典。只产出一份已复核的删词账本,不切媒体、不做字幕、不做分镜动画。用户说剪口播、处理口误、生成口播基础素材、继续剪口播,或确认卡回传 action=returncutreview 时使用。不要用于执行物理剪切、导出剪后视频、单独安装、单独打开工作台或口播分镜成片。.