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 skills add jis60224/ai-comic-skills --skill ai-comic-visual-assetsgit clone --depth 1 https://github.com/jis60224/ai-comic-skillsWrote 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/jis60224/ai-comic-skills/ai-comic-visual-assets)<a href="https://agentmods.dev/skills/jis60224/ai-comic-skills/ai-comic-visual-assets"><img src="https://agentmods.dev/badge/skills/jis60224/ai-comic-skills/ai-comic-visual-assets/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jis60224/ai-comic-skills/ai-comic-visual-assets"><img src="https://agentmods.dev/badge/skills/jis60224/ai-comic-skills/ai-comic-visual-assets.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00297 | $0.04416 |
| Opus 5 | $0.00148 | $0.02208 |
| Sonnet 5 | $0.00059 | $0.00883 |
| Haiku 4.5 | $0.00030 | $0.00442 |
Grade A, and why
ai-comic-visual-assets 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 217 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI 漫剧视觉资产提示词生成技能
本技能是 ai-comic-script（剧本生成）的下游环节：剧本写好之后，把剧本里最重要的人物、场景和道具，转成一套可复用的 AI 生图提示词，帮用户搭建自己的视觉资产库。
目标只有四句话：人物要稳定，场景要统一，道具要清楚，画风要一致。 这一步做好了，后面做分镜、视频生成、剪辑成片才不会越做越乱。
核心原则
- 所有回复使用中文。
- 母版思维：每个人物、场景、道具都先做一个"母版"提示词。母版是后续所有镜头的一致性锚点——人物不变脸、场景不换调、道具不走样，靠的就是反复参考母版。
- 全剧画风统一：先确定一个画风前缀（如"写实3D风，民国悬疑氛围"），之后所有人物、场景、道具提示词共用。不要女主写实3D、男主日系二次元。
- 先信息表，后提示词：提示词写不好，多半是信息没想清楚。先把记录表填完整并让用户确认，再转成提示词。
- 负面约束必带：每条提示词都要附负面约束，这是保证出图稳定的关键。
- 所有产出落盘：提示词、信息表、命名规范都写入文件，不要只在对话里输出。
- 本技能产出提示词，不直接生成图片：默认引导用户把提示词复制到 AI 绘图工具（即梦、Midjourney、Vidu 参考生图等）。如果当前环境有可用的图像生成工具且用户同意，也可以代为调用。
工作流程
阶段零：定位剧本来源（每次会话开始时执行）
- 扫描当前目录，查找 ai-comic-script 生成的剧本项目文件夹（特征：包含
00_剧本概览.md、01_核心人设.md、04_第XX集_完整剧本.md等文件）。 - 找到剧本项目时，读取：
00_剧本概览.md—— 剧名、题材、受众（决定画风和氛围词）01_核心人设.md—— 角色身份、性格、人物关系04_第XX集_完整剧本.md—— 默认第1集，用户指定则用指定集数05_关键画面清单.md（如有）—— 直接提示核心场景和关键道具
- 找到多个项目：列出项目名，让用户选择。
- 没有剧本项目：请用户提供以下材料（可以直接粘贴剧本原文，由技能自行提取）： 剧名、故事题材、核心人设、人物关系、第x集完整剧本、关键画面清单、核心场景、关键道具。 缺少其中几项没关系，剧本原文是最低要求。
- 检查是否已有视觉资产文件（续作模式）：如果项目里已有
07_视觉资产/，先读00_视觉资产总览.md，向用户汇报已完成的人物/场景/道具和既定画风，询问是继续补充、还是优化已有提示词。续作时必须沿用已确定的画风，不要重新设计；不要重新创建或覆盖07_视觉资产/内已有文件。 - 防止覆盖：续作模式下，
00_视觉资产总览.md、01_人物提示词.md、02_场景提示词.md、03_道具提示词.md都是全剧共享文件。为第2集及后续集数补充素材时，只能追加新角色/新场景/新道具段落，或在用户明确要求优化某个既有素材时局部修改对应段落；禁止整文件重写覆盖第1集内容。 - 补齐待补素材：如果项目里已有
08_分镜脚本/或09_视频提示词/，搜索其中的待补素材_*。当用户要求补素材、补齐待补资产、继续制作视频素材时，把这些待补项按人物/场景/道具/屏幕特写/剧情变体分类，优先生成缺口最大的资产提示词；已经在07_视觉资产/中存在的同名素材不要重复生成。
阶段一：确定全剧统一画风
根据题材向用户推荐画风并确认一次，之后全程沿用：
| 题材 | 推荐画风前缀示例 |
|---|---|
| 民国悬疑 | 写实3D风，民国悬疑氛围 |
| 现代都市 / 甜宠 | 写实3D风，现代都市氛围 |
| 古风权谋 / 玄幻 | 写实3D风，古风氛围（或国风厚涂） |
| 末世 / 科幻 | 写实3D风，末世废土氛围 |
没有特殊偏好时默认写实3D风。画风一旦确认就写进 00_视觉资产总览.md，中途更换意味着已生成的素材全部作废，所以确认这一步不能省。
阶段二：人物提示词
1. 提取核心人物，按优先级：
| 角色 | 是否必做 | 作用 | 优先级 |
|---|---|---|---|
| 主角 | 是 | 推动主线，承担主要情绪 | 高 |
| 重要对手 / 反派 | 视剧情 | 制造冲突和压迫感 | 高 |
| 核心搭档 / 情感对象 | 视剧情 | 关系拉扯、辅助推进 | 中高 |
| 本集关键配角 | 视剧情 | 提供线索 | 中 |
| 路人角色 | 否 | 氛围补充，不做母版 | 低 |
What ships with it
2 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.
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.
- 11d ago First seen · 217 lines · 297 tokens per session scan A 56b9850625ff
ai-comic-visual-assets is a skill published in the GitHub repository jis60224/ai-comic-skills (5 stars, last pushed 29d ago), licensed MIT. It adds 297 tokens to every session and 4,416 once invoked, about $0.0015 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
comic-mv-director
A production planner for animated stories, short dramas, music videos, and other AI video projects. It first decides whether the project is music-led or story-led, then plans storyboards, images, video, music, and editing.
shot-optimizer
A Chinese-language assistant for developing shots in films, animated stories, short videos, and music videos. It helps turn a rough shot idea into several prompt variations and tracks versions such as V1, V2, and V3.
story-director
A story-planning assistant that turns a rough idea into a structured storyline and a shot list. A shot list breaks a story into individual camera views, each with a purpose and visual direction.
seedance-2-guide
A reference guide to Seedance 2.0, an AI tool for generating videos from text and reference material. It explains the model's work modes, reference controls, prompt structure, and suitable video-making situations.
video-shotcraft
Create cinematic product videos from shot recipe cards, a validated template, and code/audio assets (Remotion + real page screenshots + 2.5D camera moves + beat-synced cuts + sound design). Use when the user asks to turn a frontend project or webpage into a product video, says "use video-shotcraft to make a…
shortfilm-prompt
Generate cinematic AI shortfilm prompts (works with Seedance 2.0, Xiaoyunque, Sora, Kling, Jimeng, Veo) using the 5-stage structure from Mx-Shell's Zombie Scavenger. Trigger when the user wants transformation sequences, multi-shot narrative shorts, weapon-charge/combat segments, emotional family/pet/farewell…