Brandly CLI

AI product video orchestrator — create professional product videos with AI. Works with any AI tool (OpenCode, Codex, Qwen Code, Claude Code, etc.) via CLI, agent tools, and MCP server.

🎬 What is Brandly?

Brandly helps you create professional product videos using AI — from concept to final render, all from the command line or your favorite AI assistant.

✨ What You Can Do

🎯 Who Is This For?

• Marketers & creators — make product videos without a production team
• Developers — integrate video generation into your apps/agents via CLI or MCP
• AI enthusiasts — orchestrate multi-modal AI workflows (image + video + audio)

🚀 Quick Start (5 minutes)

1. Install

ash pip install brandly-cli

2. Get an API Key

1. Sign up at Agnes AI
2. Get your API key
3. Set it as environment variable:

`ash export AGNES_API_KEY="your_key_here"

Add to your ~/.bashrc or ~/.zshrc for persistence

`

3. Create Your First Video Project

`ash

Interactive project setup

brandly init -n "My Product" -i "A sleek wireless charger that doubles as a desk lamp"

Or one-liner:

brandly init -n "Wireless Charger" -i "Minimalist 2-in-1 charger and lamp, premium materials" -s cinematic -b 200 `

4. Generate Shots (Director Mode)

`ash

Let the Director agent plan and execute

brandly run <project_id> --execute --until produce

Or step-by-step:

brandly run <project_id> --execute --until script # Write shot list brandly run <project_id> --execute --until asset # Generate shots brandly run <project_id> --execute --until validate # Quality check brandly run <project_id> --execute --until publish # Export & schedule `

5. View Results

`ash

Open visual timeline editor

brandly timeline <project_id>

Or check status

brandly status <project_id> `

📋 Core Workflows

For Creators (Human-Driven)

`ash

1. Create project

brandly init -n "Product Name" -i "Brief description" -s cinematic

2. Review & approve the shot plan

brandly status brandly approve script

3. Generate assets (one shot at a time, resumable)

brandly run --execute --until asset

4. Quality check

brandly gate --all-scenes

5. Assemble & export

brandly run --execute --until publish `

For AI Agents (Tool-Driven)

Brandly exposes 15 agent-callable tools via MCP server:

`ash

Start MCP server (for Cursor, Claude Code, etc.)

brandly mcp serve

Or use tools directly:

brandly tools --json # List all available tools brandly tools call produce_shots --project_id --shots 5 brandly tools call gate_scene --project_id --scene_id S01 `

Available tools: init_project, produce_shots, gate_scene, stitch_video, export_platforms, publish_video, estimate_cost, check_quota, egister_reference, ecover_job, and more.

🎨 Common Commands Cheat Sheet

Project Setup

ash brandly init -n "Name" -i "Idea" -s cinematic -b 500 # New project brandly status <id> # Full project overview brandly approve <id> <phase> # Human gate approval

Generation

ash brandly run <id> --execute --until produce # Full auto-pipeline brandly run <id> --execute --only shot-03 shot-07 # Regenerate specific shots brandly produce <id> --shots shots.json --interval 60 # Manual shot-by-shot brandly video <id> -p "prompt" --style commercial # Single video clip

Quality & Validation

ash brandly gate <id> --all-scenes --json # All scenes quality check brandly gate <id> --element ./clip.mp4 --ref ./ref.png # Single element check brandly validate <id> # Virality scoring

Post-Production

ash brandly stitch clip1.mp4 clip2.mp4 --ratio 2.39:1 --fit crop --output final.mp4 brandly export <id> --platforms tiktok,instagram,youtube brandly thumbnail <id> --style cinematic brandly voice-match <video> --source en --target es # Dubbing brandly beat-sync <video> <music> # Cut to beats

Publishing

ash brandly publish <id> --dry-run # Preview what will happen brandly publish <id> --platforms tiktok,instagram # Schedule posts brandly config set tiktok <token> # Store credentials

Utilities

ash brandly estimate --style cinematic --shots 8 # Cost preview brandly quota # Daily quota remaining brandly timeline <id> # Open visual editor brandly director # Show agent prompt

🎬 Video Generation Deep Dive

Styles Available

Reference Images (Consistency)

`ash

Use reference images for consistent characters/objects

brandly video -p "hero walks" --reference-images char.png,prop.png

Or register references first (recommended for multi-shot)

brandly reference --image char.png --subject-type character brandly reference --image prop.png --subject-type prop `

Keyframe Control

`ash

Start from first frame, end at last frame

brandly video -p "transition" --mode keyframe --first-frame start.png --last-frame end.png `

🤖 Director Agent (AI-Orchestrated)

The Director is an agent prompt that knows how to drive Brandly end-to-end:

`ash

Print the prompt for your AI tool

brandly director

Or run directly (experimental)

brandly run --execute --yes `

Director phases: rends → concept → script → sset → udio → e_edit → alidate → publish

Each phase has:

• Defined inputs/outputs (see randly plan --json)
• Gate command for verification
• Cost estimate before execution

📊 Monitoring & Debugging

`ash

Project status dashboard

brandly status

Shot list with progress

brandly scenes --json

Quota & budget

brandly quota brandly cost --breakdown

Job recovery (if process crashed)

brandly job-resume <video_id> brandly job-poll <video_id> --output recovered.mp4

Visual timeline editor

brandly timeline --port 8765 `

⚙️ Configuration

`ash

Set provider credentials (stored securely)

brandly config set tiktok "access_token=..." brandly config set instagram "access_token=..." brandly config set youtube "client_id=... client_secret=..."

View config

brandly config show

Environment variables

export AGNES_API_KEY="..." # Required for video/image export BRANDLY_IMAGE_CONVERT=off # Disable auto image compression export BRANDLY_WEB_PORT=8765 # Timeline editor port `

🏗️ Architecture (For Developers)

🔧 Development

`ash

Install dev dependencies

pip install -e ".[dev]"

Run tests

make test # Full suite (Python + web contracts) make lint # ruff + oxlint + mypy + import-linter

Web UI development

cd web && npm run dev

Build wheel

pip wheel . -w dist/ `

⭐ Star This Repo

If Brandly helps you create better product videos, please consider starring the repository on GitHub — it helps others discover the project and motivates continued development.

📄 License

MIT — Dream Pixels Forge