sora

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Sora Video Generation Skill

Sora视频生成Skill

Creates or manages short video clips for the current project (product demos, marketing spots, cinematic shots, UI mocks). Defaults to

sora-2

and a structured prompt augmentation workflow, and prefers the bundled CLI for deterministic runs. Note:

$sora

is a skill tag in prompts, not a shell command.

为当前项目创建或管理短视频片段（产品演示、营销广告、电影镜头、UI模拟）。默认使用

sora-2

模型和结构化提示词增强工作流，且优先使用捆绑的CLI以确保运行结果可预测。注意：

$sora

是提示词中的skill标签，而非Shell命令。

When to use

使用场景

Generate a new video clip from a prompt
Remix an existing video by ID
Poll status, list jobs, or download assets (video/thumbnail/spritesheet)
Batch runs (many prompts or variants)

从提示词生成新的视频片段
通过ID重制现有视频
轮询状态、列出任务或下载资源（视频/缩略图/精灵图）
批量处理（多个提示词或变体）

Decision tree (create vs remix vs status/download vs batch)

决策树（创建 vs 重制 vs 状态查询/下载 vs 批量处理）

If the user has a video id and wants a change → remix
If the user has a video id and wants status or assets → status/poll/download
If the user needs many prompts/assets → create-batch
If the user asks for two versions with a small change (same shot, different subject/detail) → create the base, then remix for the variation
Otherwise → create (or create-and-poll if they need a ready asset in one step)

如果用户提供了视频ID并要求修改 → 重制
如果用户提供了视频ID并要求查询状态或获取资源 → 状态查询/轮询/下载
如果用户需要处理多个提示词/资源 → 批量创建
如果用户要求生成两个仅有细微差异的版本（相同镜头，不同主体/细节） → 先创建基础版本，再重制生成变体
其他情况 → 创建（如果用户希望一步获取可用资源，则使用创建并轮询）

Workflow

工作流

Decide intent: create vs remix vs status/download vs batch.
Collect inputs: prompt, model, size, seconds, and any input reference image.
If batch: write a temporary JSONL under tmp/ (one job per line), run once, then delete the JSONL.
Prefer CLI augmentation flags (
```
--use-case
```
,
```
--scene
```
,
```
--camera
```
, etc.) instead of pre-writing a structured prompt. If you already produced a structured prompt file, pass
```
--no-augment
```
to avoid double-wrapping.
Run the bundled CLI (
```
scripts/sora.py
```
) with sensible defaults (see references/cli.md). For long prompts, prefer
```
--prompt-file
```
to avoid shell-escaping issues; pair with
```
--no-augment
```
if the prompt is already structured.
For async jobs, poll until complete (or use create-and-poll).
Download assets (video/thumbnail/spritesheet) and save locally.
Remove intermediate files created during invocation (for example
```
prompt.txt
```
,
```
remix_job.json
```
, temp JSONL). If the sandbox blocks
```
rm
```
, skip cleanup or truncate the files without surfacing an error.
Iterate with a single targeted change per prompt.

确定意图：创建、重制、状态查询/下载或批量处理。
收集输入信息：提示词、模型、尺寸、时长，以及任何输入参考图片。
若是批量处理：在tmp/目录下写入临时JSONL文件（每行一个任务），运行一次后删除该JSONL文件。
优先使用CLI增强标志（
```
--use-case
```
、
```
--scene
```
、
```
--camera
```
等），而非预先编写结构化提示词。如果已生成结构化提示词文件，需传递
```
--no-augment
```
参数以避免重复包装。
使用捆绑的CLI（
```
scripts/sora.py
```
）并采用合理默认值（详见references/cli.md）。对于长提示词，优先使用
```
--prompt-file
```
参数以避免Shell转义问题；若提示词已结构化，需搭配
```
--no-augment
```
参数。
对于异步任务，需轮询直至完成（或使用创建并轮询功能）。
下载资源（视频/缩略图/精灵图）并保存到本地。
删除调用过程中生成的中间文件（例如
```
prompt.txt
```
、
```
remix_job.json
```
、临时JSONL文件）。若沙箱环境阻止
```
rm
```
命令，可跳过清理步骤或截断文件，无需向用户报错。
每次提示词仅进行一项针对性修改，逐步迭代。

Authentication

身份验证

```
OPENAI_API_KEY
```
must be set for live API calls.

If the key is missing, give the user these steps:

Create an API key in the OpenAI platform UI: https://platform.openai.com/api-keys
Set
```
OPENAI_API_KEY
```
as an environment variable in their system.
Offer to guide them through setting the environment variable for their OS/shell if needed.

Never ask the user to paste the full key in chat. Ask them to set it locally and confirm when ready.

调用实时API必须设置
```
OPENAI_API_KEY
```
环境变量。

如果密钥缺失，请告知用户以下步骤：

在OpenAI平台UI中创建API密钥：https://platform.openai.com/api-keys
在系统中设置
```
OPENAI_API_KEY
```
为环境变量。
若需要，可指导用户根据其操作系统/Shell设置环境变量。

切勿要求用户在聊天中粘贴完整密钥。请让用户在本地设置并确认已准备就绪。

Defaults & rules

默认设置与规则

Default model:
```
sora-2
```
(use
```
sora-2-pro
```
for higher fidelity).
Default size:
```
1280x720
```
.
Default seconds:
```
4
```
(allowed: "4", "8", "12" as strings).
Always set size and seconds via API params; prose will not change them.
Use the OpenAI Python SDK (
```
openai
```
package); do not use raw HTTP.
Require
```
OPENAI_API_KEY
```
before any live API call.
If uv cache permissions fail, set
```
UV_CACHE_DIR=/tmp/uv-cache
```
.
Input reference images must be jpg/png/webp and should match target size.
Download URLs expire after about 1 hour; copy assets to your own storage.
Prefer the bundled CLI and never modify
```
scripts/sora.py
```
unless the user asks.
Sora can generate audio; if a user requests voiceover/audio, specify it explicitly in the
```
Audio:
```
and
```
Dialogue:
```
lines and keep it short.

默认模型：
```
sora-2
```
（如需更高画质，使用
```
sora-2-pro
```
）。
默认尺寸：
```
1280x720
```
。
默认时长：
```
4
```
秒（允许值：字符串类型的"4"、"8"、"12"）。
必须通过API参数设置尺寸和时长；自然语言描述不会修改这些参数。
使用OpenAI Python SDK（
```
openai
```
包）；请勿使用原始HTTP请求。
调用实时API前必须确保
```
OPENAI_API_KEY
```
已设置。
若uv缓存权限失败，设置
```
UV_CACHE_DIR=/tmp/uv-cache
```
。
输入参考图片必须为jpg/png/webp格式，且应与目标尺寸匹配。
下载URL约1小时后过期；请将资源复制到自有存储中。
优先使用捆绑的CLI，除非用户要求，否则切勿修改
```
scripts/sora.py
```
。
Sora可生成音频；若用户要求旁白/音频，需在
```
Audio:
```
和
```
Dialogue:
```
行中明确指定，并保持内容简短。

API limitations

API限制

Models are limited to
```
sora-2
```
and
```
sora-2-pro
```
.
API access to Sora models requires an organization-verified account.
Duration is limited to 4/8/12 seconds and must be set via the
```
seconds
```
parameter.
The API expects
```
seconds
```
as a string enum ("4", "8", "12").
Output sizes are limited by model (see
```
references/video-api.md
```
for the supported sizes).
Video creation is async; you must poll for completion before downloading.
Rate limits apply by usage tier (do not list specific limits).
Content restrictions are enforced by the API (see Guardrails below).

仅支持
```
sora-2
```
和
```
sora-2-pro
```
模型。
访问Sora模型API需要经过组织验证的账户。
时长限制为4/8/12秒，必须通过
```
seconds
```
参数设置。
API要求
```
seconds
```
为字符串枚举值（"4"、"8"、"12"）。
输出尺寸受模型限制（支持的尺寸详见
```
references/video-api.md
```
）。
视频创建为异步操作；下载前必须轮询直至任务完成。
存在基于使用层级的速率限制（无需列出具体限制数值）。
API会执行内容限制（详见下方安全准则）。

Guardrails (must enforce)

安全准则（必须执行）

Only content suitable for audiences under 18.
No copyrighted characters or copyrighted music.
No real people (including public figures).
Input images with human faces are rejected.

仅生成适合18岁以下受众的内容。
不得使用受版权保护的角色或音乐。
不得包含真实人物（包括公众人物）。
包含人脸的输入图片会被拒绝。

Prompt augmentation

提示词增强

Reformat prompts into a structured, production-oriented spec. Only make implicit details explicit; do not invent new creative requirements.

Template (include only relevant lines):

Use case: <where the clip will be used>
Primary request: <user's main prompt>
Scene/background: <location, time of day, atmosphere>
Subject: <main subject>
Action: <single clear action>
Camera: <shot type, angle, motion>
Lighting/mood: <lighting + mood>
Color palette: <3-5 color anchors>
Style/format: <film/animation/format cues>
Timing/beats: <counts or beats>
Audio: <ambient cue / music / voiceover if requested>
Text (verbatim): "<exact text>"
Dialogue:
<dialogue>
- Speaker: "Short line."
</dialogue>
Constraints: <must keep/must avoid>
Avoid: <negative constraints>

Augmentation rules:

Keep it short; add only details the user already implied or provided elsewhere.
For remixes, explicitly list invariants ("same shot, change only X").
If any critical detail is missing and blocks success, ask a question; otherwise proceed.
If you pass a structured prompt file to the CLI, add
```
--no-augment
```
to avoid the tool re-wrapping it.

将提示词重新格式化为结构化、面向生产的规范。仅将隐含细节明确化；不得新增创意要求。

模板（仅包含相关行）：

Use case: <where the clip will be used>
Primary request: <user's main prompt>
Scene/background: <location, time of day, atmosphere>
Subject: <main subject>
Action: <single clear action>
Camera: <shot type, angle, motion>
Lighting/mood: <lighting + mood>
Color palette: <3-5 color anchors>
Style/format: <film/animation/format cues>
Timing/beats: <counts or beats>
Audio: <ambient cue / music / voiceover if requested>
Text (verbatim): "<exact text>"
Dialogue:
<dialogue>
- Speaker: "Short line."
</dialogue>
Constraints: <must keep/must avoid>
Avoid: <negative constraints>

增强规则：

保持简洁；仅添加用户已暗示或在其他地方提供的细节。
对于重制任务，需明确列出不变项（例如“相同镜头，仅修改X”）。
若缺少关键细节导致无法成功执行，需向用户提问；否则继续执行。
若将结构化提示词文件传递给CLI，需添加
```
--no-augment
```
参数以避免工具重复包装。

Examples

示例

Generation example (single shot)

生成示例（单个镜头）

Use case: product teaser
Primary request: a close-up of a matte black camera on a pedestal
Action: slow 30-degree orbit over 4 seconds
Camera: 85mm, shallow depth of field, gentle handheld drift
Lighting/mood: soft key light, subtle rim, premium studio feel
Constraints: no logos, no text

Use case: product teaser
Primary request: a close-up of a matte black camera on a pedestal
Action: slow 30-degree orbit over 4 seconds
Camera: 85mm, shallow depth of field, gentle handheld drift
Lighting/mood: soft key light, subtle rim, premium studio feel
Constraints: no logos, no text

Remix example (invariants)

重制示例（不变项）

Primary request: same shot and framing, switch palette to teal/sand/rust with warmer backlight
Constraints: keep the subject and camera move unchanged

Primary request: same shot and framing, switch palette to teal/sand/rust with warmer backlight
Constraints: keep the subject and camera move unchanged

Prompting best practices (short list)

提示词最佳实践（简短列表）

One main action + one camera move per shot.
Use counts or beats for timing ("two steps, pause, turn").
Keep text short and the camera locked-off for UI or on-screen text.
Add a brief avoid line when artifacts appear (flicker, jitter, fast motion).
Shorter prompts are more creative; longer prompts are more controlled.
Put dialogue in a dedicated block; keep lines short for 4-8s clips.
State invariants explicitly for remixes (same shot, same camera move).
Iterate with single-change follow-ups to preserve continuity.

每个镜头仅包含一个主要动作和一个相机运动。
使用计数或节拍表示时间（例如“两步，停顿，转身”）。
若包含UI或屏幕文本，需保持文本简短并锁定相机。
当出现 artifacts（闪烁、抖动、快速运动）时，添加简短的避免项。
较短的提示词更具创意；较长的提示词控制性更强。
将对话放在专用块中；对于4-8秒的片段，需保持台词简短。
重制时需明确说明不变项（相同镜头、相同相机运动）。
通过单次修改的后续请求逐步迭代，以保持连续性。

Guidance by asset type

按资源类型提供的指导

Use these modules when the request is for a specific artifact. They provide targeted templates and defaults.

Cinematic shots:
```
references/cinematic-shots.md
```
Social ads:
```
references/social-ads.md
```

当请求特定产物时，使用以下模块。它们提供针对性的模板和默认值。

电影镜头：
```
references/cinematic-shots.md
```
社交广告：
```
references/social-ads.md
```

CLI + environment notes

CLI与环境说明

CLI commands + examples:
```
references/cli.md
```
API parameter quick reference:
```
references/video-api.md
```
Prompting guidance:
```
references/prompting.md
```
Sample prompts:
```
references/sample-prompts.md
```
Troubleshooting:
```
references/troubleshooting.md
```
Network/sandbox tips:
```
references/codex-network.md
```

CLI命令及示例：
```
references/cli.md
```
API参数速查：
```
references/video-api.md
```
提示词指导：
```
references/prompting.md
```
示例提示词：
```
references/sample-prompts.md
```
故障排除：
```
references/troubleshooting.md
```
网络/沙箱提示：
```
references/codex-network.md
```

Reference map

参考地图

references/cli.md
: how to run create/poll/remix/download/batch via
```
scripts/sora.py
```
.
references/video-api.md
: API-level knobs (models, sizes, duration, variants, status).
references/prompting.md
: prompt structure and iteration guidance.
references/sample-prompts.md
: copy/paste prompt recipes (examples only; no extra theory).
references/cinematic-shots.md
: templates for filmic shots.
references/social-ads.md
: templates for short social ad beats.
references/troubleshooting.md
: common errors and fixes.
references/codex-network.md
: network/approval troubleshooting.

references/cli.md
：如何通过
```
scripts/sora.py
```
运行创建/轮询/重制/下载/批量处理。
references/video-api.md
：API级别的参数（模型、尺寸、时长、变体、状态）。
references/prompting.md
：提示词结构与迭代指导。
references/sample-prompts.md
：可复制粘贴的提示词模板（仅示例；无额外理论）。
references/cinematic-shots.md
：电影镜头模板。
references/social-ads.md
：短社交广告节拍模板。
references/troubleshooting.md
：常见错误与修复方法。
references/codex-network.md
：网络/审批故障排除。