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 62656456/ai-film-skills --skill character-assetgit clone --depth 1 https://github.com/62656456/ai-film-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/62656456/ai-film-skills/character-asset)<a href="https://agentmods.dev/skills/62656456/ai-film-skills/character-asset"><img src="https://agentmods.dev/badge/skills/62656456/ai-film-skills/character-asset/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/62656456/ai-film-skills/character-asset"><img src="https://agentmods.dev/badge/skills/62656456/ai-film-skills/character-asset.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 | $0.00104 | $0.04198 |
| Opus 5 | $0.00052 | $0.02099 |
| Sonnet 5 | $0.00021 | $0.00840 |
| Haiku 4.5 | $0.00010 | $0.00420 |
Grade A, and why
character-asset 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 today.
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 — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
人物资产参考图生成预设
当前包已经包含全部运行规则、参数依据和质量门;不使用未经用户确认的人脸或生物特征。
0. 加载规则与映射表
确认资产设计任务 → 通读已提供剧本并建立设计依据 → 选择交付与成像模式 → 读本包画面关系判断 → 提取第1节字段 → 按第2节一致性锁定 → 第3–5节补足材质/光影/几何 → 选择平台和第7节适用模板 → 按当前交付阶段执行第8节检查与第9节负向 → 输出资产合同。资产编号:C01–C99,已确认编号不得复用。
| 节 | 独立输出 | 用途 |
|---|---|---|
| 1–5 | 资产描述、锁定规则 | 确定外观与状态 |
| 6–7 | 平台任务与提示词字段 | 直接生成参考图 |
| 8–9 | 验收和防崩 | 独立质量检查 |
| 10 | JSON 合同 | 供后续制作直接读取 |
| 11 | 类型适配字段 | 注入成片提示词的对应段落 |
剧本先行与设计依据
用户提供剧本时,第一步先通读所提供的完整正文,理解故事气质、时代与世界规则、人物身份和关系、剧情因果及时间线,再回到资产出现的场次设计外观、空间或状态。长剧本可分段读完并保留简短连续性记录,不能只搜资产名或读摘要就声称通读;只有片段时说明实际阅读范围,未给剧本时依据当前描述设计,不虚构已读全本。
内部保留简短设计依据:正文来源与阅读范围、相关场次、故事风格与成像方式、已确认事实、设计推定、关键未知。设计要能解释“为什么这个资产在此时此地是这个样子”;后文信息可帮助理解时代、身份与伏笔,但不能把后续损坏、受伤、天气或揭示结果提前带入当前场次。
用户与当前项目已经明确的画风优先;未指定成像方式时默认摄影写实,以可信结构、材质和光影呈现。悬疑、喜剧、奇幻等故事类型不自动等于某种画风。未提供细节或参考图时,从剧本与用途主动提出合理设计;影响剧情、身份、既有锁定事实的关键未知才需确认。一般颜色、材料、陈设、维护与使用细节可作有依据的设计推定,不一律退回待确认;推定不能冒充原文或用户已确认事实。
交付与成像模式
按用户点名的媒介选择;建立或重设计人物资产时,三视图加中景是最低完整交付:
- 提示词/设计说明:交付当前请求的完整提示词、必要锁定信息与资产合同;尚未生成的参考图用空列表并标为计划,不填写虚构图片路径或声称图片通过。
- 实际参考图:用户要求出图且有可用能力时生成所需图片;人物资产须包含第2节的三视图加中景。只要一张图时交一张四视图角色表,不能因此删成单个角度。实际看图后记录检查结果,用户批准前保持 pending_review。
- 诊断/局部修改:围绕给定图或提示词指出可见问题、原因假设和对应修正;保留未点名的身份、风格与状态,不默认重做整套资产。
依第0节确定成像方式(无指定时默认摄影写实),再按本包 电影画面关系判断 设计画面关系与负向。此参考在设计、改写、诊断时读取;编号/文件查询不必加载。人物资产的四项视图按第2节固定,布光与材质表达服务于所选画风和清晰可读;普通肖像、表情特写或已给图片的局部修正按点名范围处理,不冒称已交完整人物资产。
1. 描述维度表
必填维度:年龄、性别、国籍/种族、体型、脸型/眉眼鼻唇/伤疤、发型、服装层次、配饰、气质标签、特殊标记;同时说明身份职业、生活经历与当前处境怎样影响体态、衣着、面部及使用痕迹。未明确的国籍、族群或伤疤不得仅为填表而杜撰,也不默认所有人物都是年轻、俊美的精修模特。
规则:按第0节区分剧本事实、设计推定与关键未知;只把用户或剧本已确认的维度写入 locked_features,待审设计不冒充已锁定。描述按从整体到局部、从外观到材质、从静态到状态组织,并对应当前剧情节点。
2. 一致性/状态规则
人物资产必须三视图加中景:正面全身、精确90度侧面全身、背面全身,以及头部至腰部附近的中景。三面全身完整保留头顶、足部与轮廓,使用可比较的尺度与自然站姿;中景清楚展示面部、神态和上身服装。四项须在任务说明、实际生成输入和验收中同时成立;一张四格或分别四张都可承载,不将文件数量混同视图数量。
四个视图保持同一人物、年龄、脸型、发际线、体型、服装、配饰、特殊标记与剧情状态;身体左右标记不随换面镜像。脸部近景、表情、俯视和交互图只作必要补充,不能替代中景或任一全身视图。非人类角色仍交正面、侧面、背面全貌与相应中景,保留其物种结构,不强改为人形。
变体规则:服装、受伤、昼夜、损坏等变化使用原编号后缀(如 C01b)并记录变化原因;不得悄然覆盖已审核版本。参考图一经通过资产审门,后续视频只可引用,不能重新发明外观。
What ships with it
4 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.
- today Changed · +22 lines · +22 tokens per session 1237a0b431a9
- 4d ago Changed · +10 lines ff1214ab9c54
- 7d ago Changed · +10 tokens per session bfd48cc2ca38
- 12d ago First seen · 166 lines · 72 tokens per session scan A c567559f4482
character-asset is a skill published in the GitHub repository 62656456/ai-film-skills (17 stars, last pushed yesterday), licensed Apache-2.0. It adds 104 tokens to every session and 4,198 once invoked, about $0.0005 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
seedance-20
This skill should be used when creating, improving, or troubleshooting Seedance 2.0 video on any surface — Dreamina, Jimeng, CapCut, Doubao, Volcengine/Ark, BytePlus, Runway's Seedance route, fal, or third-party provider/router surfaces such as EvoLink, OpenRouter, Kie.ai, PiAPI, LaoZhang, Runware, ModelsLab, AI/ML…
character-design
A design aid for creating and revising characters used in AI-generated images or video.
wes-anderson-perspective
A visual storytelling framework based on Wes Anderson’s style, including symmetry, staged scenes, dry humour, strong colour palettes, and miniature-like compositions.
genhasamu-perspective
A creative design framework based on Kenya Hara's minimalist approach. It helps an agent examine visual ideas through simplicity, everyday objects, materials, and the use of empty space.
helmut-lang-perspective
A visual-design framework based on Helmut Lang’s minimalist fashion. It focuses on structure, technical materials, precise details, architecture, and the idea that design can be expressive without being obvious.
kawakubo-perspective
A fashion and visual-design framework based on Rei Kawakubo and Comme des Garçons. It examines deconstruction, unusual shapes, black clothing, unfinished forms, and questions about the relationship between the body and clothing.