creative-deep-dive

Orchestrates mechanism-level creative diagnostics to explain ad performance and recommend next steps.

3|Updated Apr 11, 2026
One-click install
npx skills add https://github.com/Motion-Creative/runneth-apps --skill creative-deep-dive
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: creative-deep-dive
Source: https://github.com/Motion-Creative/runneth-apps/tree/main/creative-deep-dive
Command: npx skills add https://github.com/Motion-Creative/runneth-apps --skill creative-deep-dive

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires brand-audit, paid-strategy-audit.

What problem does it solve?

creative-deep-dive solves the problem of not knowing why an ad performs the way it does, so teams can stop guessing and instead make mechanism-level, executable improvements.

Core Features & Use Cases

  • Mechanism-level deep dives (orchestrated): Runs an execution + diagnostic creative-strategy workflow to produce one coherent teardown across hook, mechanic, format, and (if applicable) body structure.
  • Standalone diagnostic skills too: Installs 9 underlying diagnostic skills as independently callable capabilities for narrower tasks like scoring a hook or identifying a tactic.
  • Grounded in upstream strategy: Reads brand-audit outputs from the workspace to avoid re-deriving strategic context on every creative review.
  • Single-creative and test-set modes: Produces per-creative deep-dive files and, for test sets, writes cross-creative comparisons that identify which test variables actually correlate with winning.

Quick Start

Ask the sandbox to run a deep dive on a creative ID and request actionable iteration recommendations based on the resulting mechanism analysis.

Frequently Asked Questions about creative-deep-dive

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

FAQPage Schema
What is mechanism-level creative diagnostics for ad teardowns?

Ad teardowns analyze hook, mechanic, format, and body structure to explain why a specific ad works or fails. This mechanism-level diagnostic approach produces structured outputs for actionable iteration rather than surface-level guesses.

How do I run a creative deep dive on a specific ad?

Request a deep dive on a specific creative ID. The workflow orchestrates underlying diagnostic skills to produce a coherent teardown across hook, mechanic, and format, delivering actionable iteration recommendations based on the mechanism analysis.

Can I compare multiple ads in a test set to find winning variables?

Yes, test-set mode supports cross-creative comparisons. It analyzes multiple creatives to identify which specific test variables actually correlate with winning performance, generating structured comparison files for review.

Do I need a brand audit before analyzing my ad creatives?

Yes, a brand audit is required to ground the strategic layer. The diagnostic reads upstream brand-audit outputs from the workspace to avoid re-deriving strategic context during every creative review, ensuring analysis aligns with brand strategy.

Can I score just the hook of an ad without a full teardown?

Yes, standalone diagnostic skills are available as independently callable capabilities. You can execute narrower tasks like scoring a hook or identifying a specific tactic without running the full orchestrated teardown workflow.

What is the best way to stop guessing why an ad fails?

The best way to stop guessing is running mechanism-level creative diagnostics. By orchestrating diagnostics across hook, mechanic, and format, you receive structured outputs explaining exactly what to change next for executable improvements.