<< All versions

Skill v1.0.0

currentLLM-judged scan90/100
pipixia-labs/creative-claw/creative-qc
──Details
PublishedOctober 23, 2026 at 05:28 AM
Content Hashsha256:8c067079d476d125...
Git SHA73d7fb045305
──Files
Files (1 file, 3.3 KB)
SKILL.md3.3 KBactive
SKILL.md · 117 lines · 3.3 KB

version: "1.0.0" name: creative-qc description: Use when the user wants to review or sanity-check a storyboard, prompt pack, script, or generated asset set for consistency, platform fit, and creative quality before final delivery or generation.


Creative QC

Use this skill to review creative work before handoff, generation, or publishing.

Trigger Cues

  • "review this storyboard"
  • "sanity-check these prompts"
  • "is this ready to generate"
  • "spot issues before we publish"
  • "quality-check this creative pack"

Best For

  • storyboard review
  • prompt pack review
  • script cleanup
  • checking generated assets against plan
  • catching mismatch before spending more generation budget

When To Use

  • The user wants review, risk detection, or final sanity-checking.
  • There is already a concrete artifact to inspect.
  • You need to compare plan vs execution, or prompt vs output.

When Not To Use

  • The user wants first-pass ideation or prompt creation, not review.
  • There is no concrete artifact yet.
  • The user already asked for direct rewriting instead of review-first feedback.

Workflow

  1. Identify the artifact type:
  • brief
  • storyboard
  • prompt pack
  • generated images
  • generated videos
  1. Run structural checks:
  • language consistency
  • count alignment
  • one-to-one mapping between scenes and prompts
  • hook strength and opening variety
  • CTA presence when the content is promotional
  • platform fit such as 9:16, 16:9, short-form pacing, or still-image use
  1. Run creative checks:
  • visual coherence
  • tone consistency
  • prompt specificity
  • repeated or generic phrasing
  1. If real assets are provided, compare them against the plan:
  • ImageUnderstandingAgent for images
  • VideoUnderstandingExpert for videos
  1. Return prioritized issues and concrete fixes.
  2. Only rewrite automatically if the user asks for direct revision. Otherwise review first.

Severity Rules

  • pass: no material blockers, only optional polish
  • revise: usable direction but there are issues that should be fixed before generation or handoff
  • block: high-risk mismatch, missing inputs, or structural failure that should stop the next step

Handoff Examples

If a prompt pack needs source comparison first:

json
{"input_path":"inbox/reference.png","mode":"prompt"}

If a generated video needs intent comparison:

json
{"input_path":"inbox/output.mp4","mode":"shot_breakdown"}

If the QC result says the script itself should be revised:

json
{"input_text":"<script to revise>","mode":"rewrite","constraints":"fix repetition and keep the original meaning"}

Output Contract

Return a QC summary like this:

json
{
"status": "pass",
"critical_issues": [],
"major_issues": [],
"minor_issues": [],
"fix_plan": [
"..."
],
"residual_risks": [
"..."
]
}

Guardrails

  • Distinguish objective mismatch from subjective taste.
  • Review first; do not silently rewrite the user's material unless asked.
  • If there are no issues, say what you checked so the pass result is meaningful.
  • When reviewing prompts, keep model compatibility in mind: concise, unambiguous, and medium-appropriate prompts usually beat overloaded prompts.
  • Prioritize findings that would waste generation budget, break platform fit, or create scene-to-prompt mismatch.
All versions