skills.rest
icoolworld
icoolworldCOMMUNITY
Explore all skills by this author →
GitHub

performance-analyst

Analyze video performance drivers from structured metadata and metrics.

Updated May 4, 2026
Search Tags:#automation#data-analytics#video-analysis#causal-inference#performance-review#template-generation#short-video
One-click install
npx skills add https://github.com/icoolworld/pvideo --skill performance-analyst-icoolworld
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: performance-analyst
Source: https://github.com/icoolworld/pvideo/tree/main/skills/performance-analyst
Command: npx skills add https://github.com/icoolworld/pvideo --skill performance-analyst-icoolworld

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Identifies and explains the drivers behind short-video performance by parsing structured data to reveal causal factors for viral and non-viral outcomes.

Core Features & Use Cases

  • Structured data analysis: Convert video_meta, production_artifacts, and performance_metrics into unified diagnostic cards.
  • Pattern recognition & causality: Trace data patterns to root causes and propose 1-2 actionable factors per video + templates for iteration.
  • Templates & iteration guidance: Generate reusable patterns and recommended template weight updates for ongoing optimization.

Quick Start

Input video_meta, production_artifacts, and performance_metrics to generate a single-video or weekly/monthly diagnostic report with actionable optimization suggestions.

Frequently Asked Questions about performance-analyst

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

FAQPage Schema
How do I identify the drivers behind short-video performance using structured data?▼

Short-video performance drivers are identified by analyzing video_meta, production_artifacts, and performance_metrics to trace data patterns to root causes. This process reveals causal factors for viral and non-viral outcomes, converting raw data into unified diagnostic cards.

What is the best way to run a weekly or monthly batch performance review for short videos?▼

The best way to run a batch performance review is inputting accumulated video_meta, production_artifacts, and performance_metrics into a structured data analyzer. This generates weekly or monthly diagnostic reports with actionable optimization suggestions across multiple video tracks.

Can I apply causal inference to single video data to generate iteration templates?▼

Causal inference can be applied to single video data to generate iteration templates by tracing performance_metrics back to production_artifacts. This isolates 1-2 actionable factors per video and outputs recommended template weight updates for ongoing optimization.

Does video analysis work for different content tracks like emotion, knowledge, and story domains?▼

Video analysis works for emotion, knowledge, story, and other content tracks by parsing structured data to reveal domain-specific causal factors. It provides actionable factors and reusable templates tailored to the unique performance drivers of each track.

What data do I need to provide to generate a video performance diagnostic report?▼

To generate a video performance diagnostic report, you need to provide three data types: video_meta, production_artifacts, and performance_metrics. Supplying these structured inputs enables pattern recognition, causality reasoning, and actionable iteration suggestions.

Related Skills

Explore More Data & Analytics Skills→
hoanghd218hoanghd218
3

plan-short-video-edit

Analyze scripts, voiceovers, and SRT transcripts to generate short video production plans.

Community
Advanced
gnoviawangnoviawan
30

marketing-video

Create video marketing strategies and production plans for brands and creators.

Community
Advanced
lloydchanglloydchang
2

deep-learning-infrastructure

Automate infrastructure pattern recognition and self-healing for complex environments.

Community
Advanced
tygwantygwan
1

analytics-reporter

Visualize usage data from .codex/analytics/metrics.jsonl into charts and summaries.

Community
Intermediate