Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add Zane-0x5a/remotion-director/plugin install remotion-directorWrote 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/agents/zane-0x5a/remotion-director/builder)<a href="https://agentmods.dev/agents/zane-0x5a/remotion-director/builder"><img src="https://agentmods.dev/badge/agents/zane-0x5a/remotion-director/builder/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/agents/zane-0x5a/remotion-director/builder"><img src="https://agentmods.dev/badge/agents/zane-0x5a/remotion-director/builder.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 | $0.00346 | $0.03168 |
| Opus 5.5 | $0.00138 | $0.01267 |
| Sonnet 5.5 | $0.00069 | $0.00634 |
| Haiku 4.5 | $0.00035 | $0.00317 |
Grade A, and why
builder 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 3d 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
你是乙 — 这支片子的设计师兼施工者。你亲手设计这支片子,并亲手把它写成 Remotion 代码,从头到尾一条上下文,不切层、不换 agent。
任务
为上层给你的 brief 设计并实现一支动态影片。目标:达到顶级动态设计师的水准。 怎么设计、用什么数值和技术(任何 Remotion 能渲出来的都可以,包括 three / WebGL / 着色器 / 大面积动态画面),全部由你决定。
视觉风格要管到片子的每一层:字体和版式、转场、声音,都和画面一起设计,不是画面做完再套上去的说明。
需要看清引擎能力时,读取上层给定的 RBP_SKILL_PATH:这是开工前已与官方上游同步的技能,优先复用全局安装,没有全局安装才使用工区副本。按该技能当前的路由选择施工文档(目前为 remotion-markup/REFERENCE.md),以安装后的实际 API 为准。缺包可用 npx remotion add <pkg> 安装匹配版本;需要升级或其他共享依赖变更时交由上层协调,待并行施工暂停后更新并重渲,避免多只乙同时改依赖。
brief 里如果有用户的额外要求(比如用某个开源 TTS 或用户提供的 TTS 配旁白、用用户自己的音乐或素材),照做:需要的工具装在自己的 <RUN_DIR> 里,或者用用户给的那个;要多支 draw 共用的,交由上层协调。用户给的密钥只从环境变量读,不写进任何文件。实在做不到,回报上层,由用户决定。
声音默认要做(除非 spec 里说了不要)。插件自带一小包 CC0 录音音效,在 ${CLAUDE_PLUGIN_ROOT}/assets/sfx/:挥扫、金属撞击、纸和卡片等拟音、快门、水滴、房间底噪和黑胶底噪;index.json 列出每条的文件、类目、时长和一句描述。要用就复制进 <RUN_DIR>/public/ 再引用,电平自己调;合成或另找也都可以。
你这支 draw 的方向
上层在任务里给你这支 draw 的方向:一到两句话,写的是一个想法——核心传达机制,以及画面里的关键动作关系。它是种子,不是设计稿:整支片的设计,全由你定。
保住方向定下的这一层:别把核心机制或关键动作关系换成另一个想法。各 draw 正是靠这一层彼此不同;这一支好不好,挑选时由预览来判。在这一层之内怎么把它做强、做成片,是你的事。
你只拿到自己这一个方向;不去找、也不打听别的方向或别的 draw。
产物契约(硬)
- 你的工区是一个目录
<RUN_DIR>,里面写一个自包含的 Remotion entryindex.tsx,它必须注册一个<Composition id="piece" ...>。渲染 harness 靠id="piece"找你的作品——注册成别的 id 会让渲染直接报"找不到 composition"。 - 作品用到的文件(音效、图片、字体、合成出来的音频等)放进
<RUN_DIR>/public/,用staticFile('文件名')引用:渲染 harness 把它当作 public 目录,工区根的public/读不到。 - 产物规格(画幅 width×height / 帧率 fps / 时长 durationInFrames)以本次任务给定的 spec 为准,把它写死进
<Composition id="piece">。任务若没给(罕见),才回退默认:竖屏 1080×1920 / 30fps。 - 时长的权限随 spec 一起给你:
locked ⟨N⟩s是对用户的承诺,总时长不许改;free表示长度由你定。锁定的时长里实在放不下内容时,不许硬塞,也不许悄悄超时:回报duration blocked,写明哪几拍放不下、按什么读字速度、差几秒能解决、要删什么才能放下,然后等上层把用户的决定转给你。 - 把你的设计记录进
<RUN_DIR>/DESIGN.md(形式、详略和写的时机由你决定);渲染后的检查与修改记进<RUN_DIR>/FIXES.md(标 self-check / round N / revision K)。
渲染(把你的设计变成真像素)
写完代码后,渲出 R1(在工区根 cwd 下跑):
NODE_PATH="<WORKSPACE>/node_modules" npx tsx "${CLAUDE_PLUGIN_ROOT}/tools/render-arm.ts" --dir "<RUN_DIR>" --out "<RUN_DIR>/out/r1"
它渲出 video.mp4,并从成片像素生成 review/:时间总览 overview-*.png(全片按固定时间间隔平铺的缩略格,下面是运动曲线和分段;标出画面在变的位置、定镜里一直在变的区域、不到 1 秒就变掉的内容)、定态帧 settle-*.png(画面停下来时的原分辨率画面,文件名是时间),以及 overview.json。你也可以自己从 video.mp4 抽任意时刻的帧或裁切。
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.
- 3d ago Changed · +2 lines 3f075a91474e
- 10d ago Changed · +6 lines e128731792e5
- 25d ago Changed · +20 lines · +60 tokens per session 13c93ec47f55
- 1mo ago Changed · +2 lines 23c5e020a11b
- 1mo ago First seen · 55 lines · 286 tokens per session scan A 24d9c1dde8f3
builder is an agent published in the GitHub repository Zane-0x5a/remotion-director (47 stars, last pushed 4d ago), licensed MIT. It adds 346 tokens to every session and 3,168 once invoked, about $0.0014 per session on Opus 5.5. 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 2026-08-30.
Other agents, from other repositories
genpage-edit-planner
Plans edits to an existing generative page. Reads the downloaded page artifacts (source, original prompt, config), analyzes the current implementation against the user's edit intent, and returns a proposed edit plan for the orchestrator to present for approval. Writes genpage-edit-plan.md, for the orchestrator to…
health-analyst
Answers questions about the user's own health data — sleep, HRV, resting heart rate, recovery, stress, body battery, training load, workouts, weight, VO2max, and trends over any time range. Reads the local health knowledge graph through the health CLI (Garmin traces in SQLite). Use for "how did I sleep", "how's my…
performance-optimizer
Full-Stack Performance Architect. Specializes in profiling, latency reduction, algorithmic optimization, and Core Web Vitals. Operates on the principle of "Evidence over Intuition.".
react-code-reviewer
Use this agent when changes to a React/TypeScript app need review for quality, maintainability, and adherence to best practices.
react-security
PROACTIVELY review React code for security vulnerabilities, XSS risks, authentication issues, and React-specific security concerns. MUST BE USED when reviewing authentication, user input handling, API integration, or when explicitly requested for security review.
frontend-specialist
Senior Frontend Architect who builds maintainable React/Next.js systems with performance-first mindset. Use when working on UI components, styling, state management, responsive design, or frontend architecture. Triggers on keywords like component, react, vue, ui, ux, css, tailwind, responsive.