video-proposal

Analyze indexed video metadata to generate structured video proposals.

Updated Feb 4, 2026
One-click install
npx skills add https://github.com/lamtuanvu/agentskills --skill video-proposal
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: video-proposal
Source: https://github.com/lamtuanvu/agentskills/tree/main/skills/video-proposal
Command: npx skills add https://github.com/lamtuanvu/agentskills --skill video-proposal

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Analyzes a collection of indexed video metadata to identify themes, moods, and clip clusters, enabling automated proposal generation for short-form content.

Core Features & Use Cases

  • Analyze collection-index.json to extract per-clip tags, moods, and durations, organizing clips into meaningful categories.
  • Generate video proposals including a collection summary, hero/clips selections, and per-category ideas ready for drafting.
  • Persist outputs to video-proposals.json to support traceability, auditing, and downstream automation with CapCut.
  • Use cases include planning TikTok/Reels concepts from an existing footage library or creating a multi-category content calendar.

Quick Start

Run the analyze_and_propose.py script on your collection-index.json to generate video-proposals.json.

Frequently Asked Questions about video-proposal

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate short-form video proposals from indexed metadata?

To generate short-form video proposals from indexed metadata, run the analyze_and_propose.py script on your collection-index.json. It reads per-clip tags, moods, and durations to output a structured video-proposals.json with categories and best clips.

What video metadata is needed to automate proposal generation?

Automating proposal generation requires a collection-index.json containing indexed video metadata. The tool specifically reads per-clip tags, moods, durations, and location data to score clip quality and organize footage into meaningful categories.

Can I use this tool to plan a TikTok or Reels content calendar?

Yes, you can plan TikTok or Reels content calendars by analyzing your footage library's metadata. The tool identifies themes and clip clusters, producing category-based ideas and a highlighted mix ready for drafting short-form video concepts.

Does the video proposal output support traceability for downstream automation?

Yes, the video proposal output supports traceability and downstream automation. Results are persisted to a video-proposals.json file, containing the collection summary, hero clips, and per-category ideas for seamless integration with CapCut.

How are video clips scored for quality during the proposal analysis?

During proposal analysis, video clips are scored for quality by evaluating the indexed metadata. The script reads per-clip tags, moods, durations, and location data to evaluate and rank the best clips for the final highlighted mix.

What is the best way to organize raw footage metadata for short-form content?

The best way to organize raw footage metadata is by creating a collection-index.json. This allows the proposal script to extract per-clip tags and moods automatically, grouping clips into meaningful categories to streamline content planning.