One-click install
npx skills add https://github.com/redbananastudios/ai-library --skill approve-output
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: approve-output
Source: https://github.com/redbananastudios/ai-library/tree/main/generated/claude/skills/approve-output
Command: npx skills add https://github.com/redbananastudios/ai-library --skill approve-output

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill streamlines the process of reviewing generated image results and turning user approvals into durable project knowledge.

Core Features & Use Cases

  • Approval status tracking: Marks a job’s generated outputs as approved by updating approval_status in metadata.json.
  • Curated image selection: Copies approved images from the job’s all/ output to selected/ for easy reuse.
  • Knowledge accumulation: Records user-approved details into winning-patterns.md and updates workflow-registry.json to improve future workflow selection.
  • Optional candidate LoRA promotion: If a candidate LoRA was used, promotes it by updating its status and approval fields for future automatic selection.

Quick Start

Tell the AI which images to approve for a given job, for example: "Approve all images for job_id 12345 and note which ones look best for the packshot style."

Frequently Asked Questions about approve-output

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

FAQPage Schema
How do I mark generated image outputs as approved in an image generation workflow?▼

To mark generated image outputs as approved, you update the approval_status field in the job metadata.json. This process also copies the approved images into a selected/ folder for reuse and appends winning patterns to winning-patterns.md.

What is the best way to save approved images for reuse from a bulk generation job?▼

The best way to save approved images for reuse is by using a bulk approval process that copies them from the job's all/ output directory into a selected/ folder. This curates your preferred assets while updating the job metadata.

Can I promote a candidate LoRA automatically after approving generated images?▼

Yes, you can promote a candidate LoRA after approving generated images by updating its status and approval fields in lora-registry.json. This ensures the successful model is selected automatically for future generation tasks.

How does human-in-the-loop feedback improve future image generation workflows?▼

Human-in-the-loop feedback improves future workflows by persisting learning signals from user approvals. It appends successful details to winning-patterns.md and updates workflow-registry.json with success metrics to guide better workflow selection.

Do I need a specific job ID to review and approve generated image candidates?▼

Yes, you need a specific job ID or output path to review and approve generated image candidates. This identifier allows the system to locate the correct metadata.json and associate the approval status with the exact generation job.

What happens to workflow-registry.json when I approve specific images from a generation job?▼

When you approve specific images, workflow-registry.json is updated with success metrics based on your selections. This records the effective configurations to help optimize and improve the selection of future image generation workflows.