以观众体验为中心预审第一人称电影解说稿,并把锁定旁白编排成节奏自然、关键事实准确的画面与声音方案。用于文稿可实现性检查、叙事节拍设计、连续原片选镜、样片制作决策、用户反馈修改,以及锁定 visualedit.json 与 audiomixplan.json。

Apache-2.0Auto-check passed

Install Movie Direct

skills CLI
$ npx skills add straighttttt/movie-commentary-workflow --skill movie-direct -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install straighttttt/movie-commentary-workflow movie-direct --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/straighttttt/movie-commentary-workflow.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.codex/skills/movie-direct .claude/skills/movie-direct && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
movie-direct
GitHub stars
135
Token cost
~470 tokens
SKILL.md length
111 words
Files
4 (incl. references)
Skills in repo
8
Repo updated
First seen
Licence
Apache-2.0

At a glance

以观众体验为中心预审第一人称电影解说稿,并把锁定旁白编排成节奏自然、关键事实准确的画面与声音方案。用于文稿可实现性检查、叙事节拍设计、连续原片选镜、样片制作决策、用户反馈修改,以及锁定 visualedit.json 与 audiomixplan.json。

  • Works in 6 steps: 读取 AGENTS.md、项目… → 如果 workflow_control.status 为… → 使用字幕与视觉索引召回候选;Qwen 只负责召回。 → …
  • SKILL.md covers 读取证据, 文稿预审, 正式编导 and 样片流程, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Movie Direct is an agent skill from straighttttt/movie-commentary-workflow. 以观众体验为中心预审第一人称电影解说稿,并把锁定旁白编排成节奏自然、关键事实准确的画面与声音方案。用于文稿可实现性检查、叙事节拍设计、连续原片选镜、样片制作决策、用户反馈修改,以及锁定 visualedit.json 与 audiomixplan.json。

Its SKILL.md is about 470 tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including reference files (for example `agents/openai.yaml`, `references/directing-principles.md` and `references/plan-contract.md`).

The repository describes itself as: Auditable Codex skills and data contracts for first-person movie commentary production. The licence is Apache-2.0.

Example prompts

  • “/movie-direct”

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. 读取 AGENTS.md、项目 production/state.json、project.json、剧情地图、锁定文稿或旁白。
  2. 如果 workflow_control.status 为 PAUSED,且当前用户没有明确授权恢复,立即停止。
  3. 使用字幕与视觉索引召回候选;Qwen 只负责召回。
  4. 关键动作、对白、揭示、交接和身份变化必须回到带声音的连续原片。
  5. 设计画面前读取 references/directing-principles.md。
  6. 输出机器方案前读取 references/plan-contract.md。

What it can do on your machine

Read from SKILL.md and the folder at commit 3d50c17. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Movie Direct loads about 470 tokens when it runs, and up to ~1.3k if it reads all its reference files. Until then it costs about 36 tokens; SKILL.md has 111 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~36
When it runs · the whole SKILL.md, loaded when a task matches
~470
With references · SKILL.md plus every file in references/, read only if the agent opens them
~1.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from straighttttt/movie-commentary-workflow at commit 3d50c17, republished under its Apache-2.0 licence (© straighttttt). 111 words, ~470 tokens.

Download SKILL.mdSave it as .claude/skills/movie-direct/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
movie-direct
description
以观众体验为中心预审第一人称电影解说稿,并把锁定旁白编排成节奏自然、关键事实准确的画面与声音方案。用于文稿可实现性检查、叙事节拍设计、连续原片选镜、样片制作决策、用户反馈修改,以及锁定 visual_edit.json 与 audio_mix_plan.json。

电影解说编导

担任创意负责人。先完成好看的剪辑,再用轻量验收发现关键错误;不得把验收表当成选镜方法。

读取证据

  1. 读取 AGENTS.md、项目 production/state.json、project.json、剧情地图、锁定文稿或旁白。
  2. 如果 workflow_control.status 为 PAUSED,且当前用户没有明确授权恢复,立即停止。
  3. 使用字幕与视觉索引召回候选;Qwen 只负责召回。
  4. 关键动作、对白、揭示、交接和身份变化必须回到带声音的连续原片。
  5. 设计画面前读取 references/directing-principles.md。
  6. 输出机器方案前读取 references/plan-contract.md。

文稿预审

  1. 检查事实、第一人称获知边界、反转顺序和可用画面。
  2. 只把事实错误、身份错误、提前泄露和找不到关键画面列为阻断项。
  3. 编导不得直接覆盖执笔人正文;把目标结果退回执笔人。

正式编导

要求旁白和时间戳已经锁定。

  1. 听完整篇旁白,按约 8~15 秒划分叙事节拍。
  2. 对每个节拍判断观众最需要理解和感受的内容,不按标点机械切换。
  3. 将内容分为 A级关键事实、B级情绪心理、C级过渡空间。
  4. 为每个节拍搜索多个候选,查看连续原片后选择 1~3 个镜头。
  5. 优先保留完整表演、动作、空间和悬念;允许画面在不误导的前提下轻微领先或滞后旁白。
  6. 安排需要保留的原片对白、环境声、静默和音乐。
  7. 完成结局后反查受反转影响的早期镜头。

样片流程

  1. 输出覆盖样片时间轴的 visual_edit.json 和 audio_mix_plan.json。
  2. 只为 A级关键事实编写 anchor_map.json;普通句不强制创建逐句语义证据。
  3. 检查时间轴与小于 1 秒的后期拼接风险后,锁定代理方案。
  4. 交给 movie-render-qa 生成可播放样片。
  5. 样片通过基础技术检查后,编导正常速度完整观看一次。
  6. 只对 A级阻断错误和明显节奏问题集中修正一次;不得继续内部循环。
  7. 将通过基础技术检查的最新样片立即交给用户。
  8. 只有用户认可样片后,才能扩展全片并建立正式导演锁。

全片扩展

  1. 按已获认可的方向完成全片方案。
  2. 生成低成本全片拼接版并正常速度完整观看一次。
  3. 集中修正 A级阻断错误和明显连续性问题后,锁定正式导演方案。

原片接管叙事

把原片接管作为按内容决定的可选手段,不设置配额。

  1. 只在关键对白、表演、动作或现场声音能独立承担叙事,并且明显强于解说复述时使用。
  2. 优先替换附近表达同一事实的解说范围,避免同一信息重复两遍;不能安全替换时,宁可不插。
  3. 进入原片后完全停止解说,保留同步原声和必要字幕;在完整句子、动作或情绪结束后,再从下一句完整解说或下一节拍继续。
  4. 保留足够的连续上下文,不从关键台词中间进入或退出,不为凑时长使用普通对白。
  5. 不得提前泄露反转,也不得因为某个案例成功就复制其数量、时长或剧情位置。
  6. 在导演方案中记录被替换的解说范围、原片精确坐标、使用目的、原声与字幕要求;未明确登记的段落不得由渲染阶段自行加入。

创意判断

  • A级事实错误、错人物、错事件、错身份或反转提前属于阻断。
  • B级与C级以观众是否自然理解、情绪是否连贯为准。
  • 毫秒级轻微跨界不可见且不误导时,不判整版失败。
  • 争议段落或用户点名问题可以生成局部证据。

保护创作权

  • 不让检索排名、PySceneDetect 或渲染模块决定镜头。
  • 不用两分钟媒体包代替剧情结构。
  • 不因时间接近就接受错误镜头。
  • 不让内部验收阻止用户看到最新可播放样片。

© straighttttt, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files (references) in .codex/skills/movie-direct of straighttttt/movie-commentary-workflow.

  • SKILL.md
  • agents/openai.yaml
  • references/directing-principles.md
  • references/plan-contract.md

Open the folder on GitHubat commit 3d50c17

Compare with similar skills

Movie Direct next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Movie Direct compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Movie Direct this skillstraighttttt/movie-commentary-workflow135—~470Automated safety check: PassApache-2.0
Direction Pickernexu-io/open-design100k—~472Automated safety check: WarnApache-2.0
Frontend Design Directionaffaan-m/ECC277k1 repos~2.2kAutomated safety check: PassMIT
Direction Attributethedaviddias/Front-End-Checklist74k—~534Automated safety check: PassMIT
Directautonomous-ai/openharness1.2k—~820Automated safety check: PassMIT
Direct Testsinternet-court/internet-court-skill6.6k1 repos~1.2kAutomated safety check: NotesMIT

Similar skills

  • Direction Picker

    nexu-io/open-design

    Resolves the visual direction at the plan stage from the brief and design system, without asking the user.

    100k GitHub stars~472 tokensUpdated yesterday
    Frontend & DesignAuto-check: warnings
  • Set an ECC-specific frontend design direction for production UI work.

    277k GitHub starsUsed in 1 repo~2.2k tokens
    Frontend & DesignAuto-check passed
  • Direction Attribute

    thedaviddias/Front-End-Checklist

    A skill your agent uses when reviewing templates, rendered HTML, or shared components related to Set text direction for RTL languages.

    74k GitHub stars~534 tokensUpdated 5 days ago
    Auto-check passed
  • Direct

    autonomous-ai/openharness

    Create and revise a usable brand identity and coordinated launch materials from a business brief, using the Forme project and the person's own artwork, text and production requirements.

    1.2k GitHub stars~820 tokensUpdated today
    Media & CreativeAuto-check passed
  • Direct Tests

    internet-court/internet-court-skill

    Write and run fast direct mode tests for GenLayer intelligent contracts.

    6.6k GitHub starsUsed in 1 repo~1.2k tokens
    Testing & QAAuto-check: notes
  • Use direct Python execution for reliable spreadsheet operations instead of shellagent

    7.8k GitHub stars~1.2k tokensUpdated 1 mo ago
    Documents & OfficeAuto-check passed

More from straighttttt/movie-commentary-workflow

All 8 skills in this repo
  • Movie Voice Tts

    straighttttt/movie-commentary-workflow

    使用千问 Audio 3.0 TTS Plus、MiniMax Speech 2.8 HD 和 Fish Audio S2.1 Pro 生成或克隆电影角色旁白,包括直接复用用户选定的 Fish 公共音色编号并保存原生时间戳;供应商没有原生时间戳时,使用火山引擎语音识别恢复中文字幕或字级时间戳。制作中英文第一人称电影旁白、三模型克隆试音、供应商对比、整篇配音、字幕定时或音文对齐时使用。

    135 GitHub stars~1.1k tokensUpdated 2 mo ago
    Auto-check: notes
  • Movie Second Review

    straighttttt/movie-commentary-workflow

    独立完整复看已经渲染的电影解说成片,逐秒检查最终观众实际看到和听到的内容,定位瞬闪、秒切、跳画面、残帧、节奏突变、字幕与声音接缝,并把创意问题以精确时间码和证据交回 movie-direct。用户要求全片复检、二次审片、检查快切或跳帧、对成片提出时间码修改意见,或者在 FINALRENDERED 后决定能否进入 FINALVERIFIED 时使用。

    135 GitHub stars~462 tokensUpdated 2 mo ago
    Auto-check passed
  • Movie Master Director

    straighttttt/movie-commentary-workflow

    让同一个主脑在不同 Codex 窗口中接管并连续完成一部电影解说,从原片索引、全片理解、第一人称文稿、角色配音、分批选镜、渲染到正式复检,同时通过项目状态、导演记忆和认可样片保持创意连续性。开始新电影、接管或续做现有电影、查询当前制作阶段、处理用户反馈、制作样片或成片时使用;默认作为所有电影解说项目的唯一总入口。

    135 GitHub stars~545 tokensUpdated 2 mo ago
    Auto-check passed
  • Movie First Person Writer

    straighttttt/movie-commentary-workflow

    理解整部电影,并撰写可直接进入制作的第一人称电影解说稿,严格控制主角的主观信息边界、心理推进、反转揭露时机和编导审核版本。选择第一人称叙述者、建立剧情地图或主角简报、起草或整体重写文稿、修正事实与视角问题,或者在配音前锁定文稿时使用。

    135 GitHub stars~407 tokensUpdated 2 mo ago
    Auto-check passed
  • Movie Index

    straighttttt/movie-commentary-workflow

    将本地电影原片、字幕、技术切点、关键帧和视觉模型观察结果整理成经过验证的可搜索镜头数据库。开始新的电影解说项目、重建或检查索引、按对白或画面寻找素材、返回原片精确坐标与证据片段,或者为执笔人和编导提供共享证据时使用。

    135 GitHub stars~389 tokensUpdated 2 mo ago
    Auto-check passed
  • Movie Render QA

    straighttttt/movie-commentary-workflow

    忠实执行已锁定的电影解说样片或正式导演方案,完成截取、拼接、字幕、混音、增量渲染和技术验收。用于快速生成可供用户判断的样片,或对用户认可后的正式成片执行完整技术检查;不得替编导改变镜头和节奏。

    135 GitHub stars~326 tokensUpdated 2 mo ago
    Auto-check passed

Questions about Movie Direct

What does Movie Direct do?

以观众体验为中心预审第一人称电影解说稿,并把锁定旁白编排成节奏自然、关键事实准确的画面与声音方案。用于文稿可实现性检查、叙事节拍设计、连续原片选镜、样片制作决策、用户反馈修改,以及锁定 visualedit.json 与 audiomixplan.json。. Movie Direct is an agent skill from straighttttt/movie-commentary-workflow.

How do I install Movie Direct in Claude Code?

Run `npx skills add straighttttt/movie-commentary-workflow --skill movie-direct -a claude-code`. Or copy the skill folder (.codex/skills/movie-direct in straighttttt/movie-commentary-workflow) into .claude/skills/movie-direct in your project. Claude Code loads it when a task matches its description.

How do I install Movie Direct in Codex?

Run `npx skills add straighttttt/movie-commentary-workflow --skill movie-direct -a codex`. Or copy the skill folder (.codex/skills/movie-direct in straighttttt/movie-commentary-workflow) into .agents/skills/movie-direct in your project. Codex loads it when a task matches its description.

Can I use Movie Direct in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add straighttttt/movie-commentary-workflow --skill movie-direct -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/movie-direct, .gemini/skills/movie-direct, .github/skills/movie-direct and .opencode/skills/movie-direct in your project.

What does Movie Direct need to run?

SKILL.md names no scripts, command-line tools or credentials: Movie Direct is instructions for the agent only.

Does Movie Direct access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Movie Direct safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Movie Direct use?

Movie Direct is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Movie Direct use?

About 470 tokens (SKILL.md is roughly 1.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 852 tokens, read only when the agent opens those files.

What are the alternatives to Movie Direct?

Skills that share tags, products or a category with Movie Direct: Direction Picker (nexu-io/open-design, 100k stars), Frontend Design Direction (affaan-m/ECC, 277k stars), Direction Attribute (thedaviddias/Front-End-Checklist, 74k stars) and Direct (autonomous-ai/openharness, 1.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Movie Direct?

straighttttt (a GitHub user) maintains it in straighttttt/movie-commentary-workflow, which has 135 GitHub stars. The repository holds 8 skills in this directory. The repository was last updated on August 11, 2026.

Source: straighttttt/movie-commentary-workflow on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.