AI Output Eval Scorecard
Score AI-generated marketing outputs across accuracy, clarity, brand fit, policy compliance, actionability, and creative operating system fit.
Use when AI-generated marketing outputs need repeatable scoring across quality, compliance, actionability, and creative operating system fit.
Inputs
- task_type
- prompt_used
- model_output
- brand_rules
- policy_rules
- scoring_weights
- optional_creative_operating_system_dimensions
Outputs
- Scorecard
- Verdict
- Rewrite recommendations
- Creative operating system scores
Before you begin
- Reading and copying the procedure require no terminal setup. Check requirements for any packaged programs separately.
Try this first
Read the procedure and supply its documented inputs. Use the full text in an agent app or chat that supports pasted instructions, following that host's guidance.
Limits for this step
- Instruction package only; it does not provide or authorize an executable runtime capability.
- Safe for evaluation and recommendation; require human approval before approving externally visible copy.
Read full instructionsHide full instructions
AI Output Eval Scorecard
When to use
Use before shipping AI-generated copy, reports, recommendations, campaign concepts, creator briefs, or creative strategy outputs when you need a repeatable quality and risk score.
Inputs required
- task_type
- prompt_used
- model_output
- brand_rules
- policy_rules
- scoring_weights
- optional_creative_operating_system_dimensions
Workflow
- Parse task context and output.
- Score core dimensions: accuracy, clarity, brand fit, policy compliance, actionability.
- If the output is campaign, creator, cultural, or Cannes-style creative work, also score Creative Operating System dimensions:
- utility: does it help people do something useful?
- product_as_media_potential: can the idea live in product, service, commerce, CRM, app, packaging, or another owned surface?
- cultural_timing: is the timing earned, safe, and relevant?
- human_meaning: does it connect to a real behavior, tension, need, or social context?
- brand_memory: does it use approved brand knowledge and avoid generic category language?
- distinctiveness: would the output still be recognizable if the logo was removed?
- creator_fit: if creators are involved, does the idea preserve creator voice and audience trust?
- Flag critical findings with severity and evidence.
- Provide targeted rewrite or rework recommendations.
- Return verdict and confidence.
Output format
- overall_score (0-100)
- verdict (
pass,revise,fail) - dimension_scores
- creative_operating_system_scores (when relevant)
- critical_findings
- rewrite_recommendations
- confidence
Guardrails
- Never infer policy pass if rules are incomplete.
- Cite concrete evidence for each high-severity finding.
- Do not fabricate legal or platform rules.
- Do not reward polished language if the idea lacks evidence, utility, or brand permission.
- Flag unsupported claims even when they sound strategically attractive.
View file metadata
---
name: ai-output-eval-scorecard
description: Evaluate and score AI marketing outputs with a consistent rubric for quality, compliance, deployment readiness, and creative operating system fit.
---Install instruction files
These commands copy guidance into the selected runtime directories. Confirm that your agent app discovers it. Any packaged programs need separate setup and verification.
Install (Codex)
./bin/skills-hub install marketing/[email protected] --runtime codex
Install (Claude)
./bin/skills-hub install marketing/[email protected] --runtime claude
Install (Generic)
./bin/skills-hub install marketing/[email protected] --runtime generic --target ./my-agent/skills