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 xixihhhh/ai-short-drama-skill --skill dramakegit clone --depth 1 https://github.com/xixihhhh/ai-short-drama-skillWrote 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/xixihhhh/ai-short-drama-skill/dramake)<a href="https://agentmods.dev/skills/xixihhhh/ai-short-drama-skill/dramake"><img src="https://agentmods.dev/badge/skills/xixihhhh/ai-short-drama-skill/dramake/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/xixihhhh/ai-short-drama-skill/dramake"><img src="https://agentmods.dev/badge/skills/xixihhhh/ai-short-drama-skill/dramake.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 | $0.00279 | $0.06117 |
| Opus 5 | $0.00139 | $0.03059 |
| Sonnet 5 | $0.00056 | $0.01223 |
| Haiku 4.5 | $0.00028 | $0.00612 |
Grade A, and why
dramake 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 — 269 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dramake
把短剧制作当成一条连续生产线,而不是一堆提示词。让没有编剧、导演或剪辑经验的用户,也能从一段文字开始,在明确的验收门禁下完成故事、剧本、分镜、生成、配音、剪辑和质检;让专业用户可以从任意阶段接管。
本 Skill 不绑定单一平台、网站、API 或 MCP,执行平台由用户选择。使用用户能合法访问并愿意使用的模型与执行入口;每次执行前核实当前参数、价格、素材上传与保存规则。
以 project-state.md 作为唯一生产状态机。提示词、接口参数和时间线都从已批准的创意状态与已检查的真实素材派生。
十条铁律
- 主角必须因果性改变结果,人、动物、产品或物体都不能只是装饰。
- 静音观看也应能理解主要因果与重点笑点;对白、字幕和群演笑声不能替代可见铺垫、原因、反应与揭晓。
- 模型支持时优先使用更少、更长的剧情段,但必须限制事件负载:段内可有多个节拍,不能让一个提示词同时承担多条高风险物理链、道具变种和无关任务。
- 未实际下载和验收上游素材,不得生成依赖其状态的下游段落。
- 锁定身份、材质、服装结构、轴线、道具、天气、湿度、光线和声音边界。
- 接触必须有接近、接触、施力、响应和稳定结果;拒绝穿模与不可能解剖。
- 默认不在视频里生成手机 UI、信件、价签、倒计时和关键文字;后期叠加。
- 原生对白可以解决口型,但每句必须锁定唯一说话者、可见嘴型与非说话者闭嘴状态;原生音乐不能自动解决终片混音。
- 技术干净不等于剧情能发;60–120 秒故事必须有贯穿全片的宏观因果链和有铺垫的主反转,并进行未读梗概的冷观众检查。
- 没有真实看过、听过或抽帧检查,就不得声称素材已通过。
从用户会什么开始
先判断用户输入深度,不要要求小白先学会专业术语:
- **一句话/脑洞:**主动给出 2–3 个可拍方向,补齐主角、欲望、阻碍、升级与结果;
- **故事/小说/文案:**抽取人物、关系、时间线、道具、秘密、转折与可视化动作;
- **剧本:**检查因果、铺垫、反转、对白密度、场景数量和时长;
- **分镜/提示词:**检查节拍、连续性、物理、文字风险和模型可执行性;
- **已有素材/粗剪:**实际检查画面与声音,定位最小失败单元,不无故推倒重来。
再选任务深度:
- **诊断/建议:**只分析,不生成、不付费;
- **快速修复:**只修剧本、提示词、镜头、接缝、声线或混音的最小单元;
- **生产包:**输出简报、剧本、母版、分镜、清单、提示词、费用与剪辑方案;
- **执行/续作:**从当前已验收状态继续;
- **完整制作:**从文字到发布执行整条状态机。
局部问题不得被强制重走整条流程。
选择内容形态与画幅
先选择当前交付的主形态,详见 format-routing.md:
- **15–45 秒独立小短片:**一个钩子、一次升级、一个回报;
- **45–120 秒独立短片:**2–4 个剧情段,完成情绪或喜剧弧;
- **1–4 分钟单元剧:**固定角色/世界规则,每集问题独立解决;
- **1–4 分钟连续短剧:**单集目标与回报之外,还维护季线状态。
支持 9:16、16:9、1:1 和自定义比例。记录发布渠道、画幅、主体安全区、字幕区、封面裁切区与交付编码。
需要横竖双版本时,共用剧情、人物、声线、道具与连续性母版,分别设计构图、景别、移动路线和安全区,分别验收。不得默认中心裁切或机械补边。
建立生产简报
使用 production-brief.md。从对话和素材推断安全默认值,只询问会实质改变结果的选择。
至少记录:
- 原始文字与改编边界;
- 渠道、画幅、时长、受众、语言和情绪承诺;
- 人物、动物、产品、地点、参考素材与使用权;
- 独立/单元/连续形式与内容配置;
- 可用模型/执行入口和大致预算上限;
- 对白策略:无对白、原生、参考音频、外部配音或混合;
- 视觉介质:真人、写实宠物、动画、漫剧、UGC 或混合。
公开示例不得暴露密钥、签名链接、请求 ID、私人路径、真实脸/声音或私有参考素材。
先写故事,再写提示词
阅读 story-and-pacing.md 与 directing-camera-and-pacing.md;喜剧、萌宠、拟人或动画化表演项目同时阅读 comedy-and-animated-performance.md。交付:
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/templates/audio-continuity.csv 368 B
- assets/templates/block-manifest.csv 685 B
- assets/templates/candidate-scorecard.csv 294 B
- assets/templates/character-bible.md 1.6 KB
- assets/templates/continuity-ledger.csv 407 B
- assets/templates/cost-plan.csv 394 B
- assets/templates/coverage-plan.csv 278 B
- assets/templates/edit-timeline.csv 243 B
- assets/templates/en/character-bible.md 1.9 KB
- assets/templates/en/production-brief.md 2.8 KB
- assets/templates/en/prop-bible.md 2.5 KB
- assets/templates/en/qa-report.md 2.7 KB
- assets/templates/generation-jobs.csv 396 B
- assets/templates/model-card.yaml 1.4 KB
- assets/templates/production-brief.md 2.6 KB
- assets/templates/prop-bible.md 2.4 KB
- assets/templates/qa-report.md 2.6 KB
- assets/templates/shot-plan.csv 733 B
- assets/templates/spatial-plan.csv 412 B
- assets/templates/spend-ledger.csv 459 B
- assets/templates/story-memory.csv 241 B
- assets/templates/voice-bible.csv 252 B
- references/audio-and-dubbing.md 5.1 KB
- references/authorization-and-batch.md 1.5 KB
- references/budget-and-quality.md 1.6 KB
- references/comedy-and-animated-performance.md 6.6 KB
- references/content-profiles.md 2.2 KB
- references/continuity-and-transitions.md 1.6 KB
- references/directing-camera-and-pacing.md 7.0 KB
- references/editing-delivery-and-release.md 2.4 KB
- references/en/audio-and-dubbing.md 6.6 KB
- references/en/authorization-and-batch.md 1.9 KB
- references/en/budget-and-quality.md 2.1 KB
- references/en/comedy-and-animated-performance.md 7.7 KB
- references/en/content-profiles.md 2.9 KB
- references/en/continuity-and-transitions.md 1.9 KB
- references/en/directing-camera-and-pacing.md 7.5 KB
- references/en/editing-delivery-and-release.md 2.7 KB
- references/en/format-routing.md 1.7 KB
- references/en/generation-blocks-and-edit-beats.md 2.8 KB
- references/en/identity-voice-and-reference-locks.md 4.8 KB
- references/en/model-router.md 8.0 KB
- references/en/physics-and-failure-modes.md 4.9 KB
- references/en/previsualization-and-coverage.md 3.1 KB
- references/en/production-operations-and-recovery.md 3.8 KB
- references/en/project-state.md 4.1 KB
- references/en/prompt-compiler.md 3.8 KB
- references/en/qa-and-repair.md 4.2 KB
- references/en/reference-graph-and-storyboard-images.md 4.1 KB
- references/en/series-memory-and-source-adaptation.md 2.2 KB
- references/en/spatial-blocking-and-world-state.md 2.9 KB
- references/en/story-and-pacing.md 6.6 KB
- references/en/style-and-medium-locks.md 4.1 KB
- references/format-routing.md 1.2 KB
- references/generation-blocks-and-edit-beats.md 2.3 KB
- references/identity-voice-and-reference-locks.md 4.5 KB
- references/model-router.md 6.6 KB
- references/physics-and-failure-modes.md 3.9 KB
- references/previsualization-and-coverage.md 2.9 KB
- references/production-operations-and-recovery.md 3.5 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.
- 11d ago First seen · 269 lines · 279 tokens per session scan A 076b67eeb101
dramake is a skill published in the GitHub repository xixihhhh/ai-short-drama-skill (15 stars, last pushed 12d ago), licensed MIT. It adds 279 tokens to every session and 6,117 once invoked, about $0.0014 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
tapcanvas-storyboard-expert
A chapter-based storyboard and video-planning guide for turning stories or scripts into production-ready shot plans. It covers characters, locations, props, camera direction, dialogue, voice-over, subtitles, and short video timelines.
seedance-2-video-gen
A reference guide for planning videos with Seedance 2.0, including text-to-video, image-to-video, and mixed image, video, and audio inputs.
tapcanvas-api
A single setup and command-line entry point for calling TapCanvas's public chat, drawing, image-understanding, video, task-result, and flow APIs.
tapcanvas-demo-patterns
A set of methods for keeping characters, settings, camera views, and visual style consistent when creating images or videos.
tapcanvas-collage-local
Use local Python Pillow to compose storyboard/role-card collages with per-cell role-name labels. No model call needed.
tapcanvas-workflow-orchestrator
A TapCanvas planning guide that uses live project, flow, node, chapter, and continuity information to decide the next canvas action. It is intended for multi-step creative workflows rather than ordinary one-off questions.