@talocode/cliploop

Turn a sentence into a video. From concept to rendered output — brief, script, render — in one pipeline.

npm install @talocode/cliploop

Why ClipLoop?

Most video creation tools make you jump between 5 different apps to get a finished piece of content. ClipLoop is different — it's a single prompt-to-video pipeline.

Describe what you want → get a structured brief → get a scene-by-scene script → submit for render. The entire creative workflow is programmable.

Who is this for?

• Developers building video generation into their products
• Content teams automating short-form video at scale
• Marketers creating campaign packages across platforms

How it works

flowchart LR
  A[Your Prompt] --> B[Brief]
  B --> C[Script]
  C --> D[Render]
  D --> E[Video]

Brief — Takes a natural language prompt and produces a structured content plan: hook, structure, tone, target audience.

Script — Turns the brief into a scene-by-scene video script with visuals, narration, and timing for each scene.

Render — Submits the script to a render engine and returns a downloadable video file.

Quick Start

# 1. Get a brief for your video idea
cliploop brief --prompt "explain transformers in 60 seconds" --channel youtube

# 2. Generate a script from the brief
cliploop script --brief-id brief_abc123 --style educational

# 3. Render the video
cliploop render --script-id script_def456 --format landscape

# 4. Check status
cliploop status --video-id render_ghi789

SDK (in your code)

import { generateBrief, generateScript, submitRender } from '@talocode/cliploop'

// Step 1: Generate a content brief
const brief = await generateBrief({
  prompt: 'explain quantum computing in 60 seconds',
  channel: 'youtube',
  duration: 60,
})
// → { id, title, hook, structure, tone, targetAudience }

// Step 2: Generate a scene-by-scene script
const script = await generateScript({
  briefId: brief.id,
  style: 'educational',
})
// → { id, scenes: [{ visual, narration, duration }] }

// Step 3: Submit for rendering
const render = await submitRender({
  scriptId: script.id,
  format: 'landscape',
  quality: 'standard',
})
// → { id, status: 'processing' }

// Step 4: Poll until complete
const status = await getRenderStatus(render.id)
// → { status: 'completed', videoUrl: '...' }

CLI Reference

| Command | Description | Required | |---------|-------------|----------| | cliploop brief | Generate a video content brief | --prompt | | cliploop script | Generate a script from a brief | --brief-id | | cliploop render | Submit a video for rendering | --script-id | | cliploop status | Check render job status | --video-id | | cliploop campaign | Create a campaign package | --name, --platform |

Options:

• --cloud — Route through Talocode Cloud API (requires TALOCODE_API_KEY)
• --format json|text — Output format

Campaigns

Bundle multiple videos into a platform-ready campaign:

cliploop campaign --name "Product Launch Q3" --platform youtube --schedule "2026-08-01T00:00:00Z"

Environment

| Variable | Required | Default | Description | |----------|----------|---------|-------------| | MISTRAL_API_KEY | For local AI | — | Mistral AI API key | | TALOCODE_API_KEY | For cloud mode | — | Talocode Cloud API key | | TALOCODE_BASE_URL | No | https://api.talocode.site | API endpoint |

License

MIT