skills.rest
ktkarchive
ktkarchiveCOMMUNITY
Explore all skills by this author →
GitHub

codex-imagegen-2-skill-for-kimi

Generate and edit images from text prompts and reference images via OpenAI Codex OAuth.

24|6|Updated Apr 22, 2026
Search Tags:#prompt engineering#image generation#image editing#codex oauth#png validation#kimi cli#gpt-5.5
One-click install
npx skills add https://github.com/ktkarchive/codex-imagegen-2-skill-for-kimi --skill codex-imagegen-2-skill-for-kimi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: codex-imagegen-2-skill-for-kimi
Source: https://github.com/ktkarchive/codex-imagegen-2-skill-for-kimi/tree/main
Command: npx skills add https://github.com/ktkarchive/codex-imagegen-2-skill-for-kimi --skill codex-imagegen-2-skill-for-kimi

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

You want high-quality image generation and image-to-image editing from natural language without managing API keys or per-image billing.

Core Features & Use Cases

  • Text-to-Image Generation: Turn a prompt into a high-resolution image using OpenAI Codex OAuth.
  • Reference Image Editing: Transform an existing image (style transfer, background swap, object replacement) using structured instructions.
  • Quality Guardrails: Validate output PNG integrity and optionally score prompt-image alignment with an automatic retry when misaligned.
  • History Logging: Record each generation/edit run with prompts, parameters, verification, and outputs.

Quick Start

Use the skill in Kimi by cloning it into your ~/.kimi/skills/ directory and then asking it to generate an image, for example: generate an image of a cat in space.

Frequently Asked Questions about codex-imagegen-2-skill-for-kimi

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

FAQPage Schema
How do I generate images from text without managing API keys?▼

You can generate images from text without API keys by using OpenAI Codex OAuth. This process proxies authentication through a local startup on port 10531, enabling tool-driven image_generation calls for high-resolution outputs.

Can I edit an existing image using text prompts and a reference image?▼

Yes, you can perform image-to-image transformations using a reference image and structured text instructions. This supports style transfer, background swaps, and object replacement without requiring manual API key configuration.

How does prompt-image alignment scoring work during image generation?▼

Prompt-image alignment scoring evaluates the generated output against your original text prompt. If the output is misaligned, the system automatically retries the generation to satisfy deterministic verification requirements.

What is the best way to validate PNG integrity for generated images?▼

The best way to validate PNG integrity is through deterministic output verification built into the generation workflow. This ensures the final image file is structurally sound and automatically logs the verification results.

Do I need per-image billing for text-to-image generation with Codex OAuth?▼

No, you do not need per-image billing for text-to-image generation. Using OpenAI Codex OAuth bypasses standard API key billing, allowing you to create high-resolution images from natural language without per-image costs.

How do I start the OAuth proxy for image generation in Kimi CLI?▼

To start the OAuth proxy for image generation in Kimi CLI, clone the skill into your ~/.kimi/skills/ directory. This initiates the proxy on port 10531, allowing you to ask Kimi to generate images directly.

Related Skills

Explore More Design & Creative Skills→
ManuelKugelmannManuelKugelmann
3

claude-get-session-info

Export CLAUDE_PID and SESSION_ID variables by sourcing a shell script.

Community
Basic
Zhow01Zhow01
52

baidu-search

Query Baidu AI Search to retrieve live web results with configurable count and freshness.

Community
Intermediate
OpenMinisOpenMinis
361

codex-image

Generate images from natural language prompts via the ChatGPT backend.

Official
Advanced
landfilllandfill

gpt-image-1-5

Generate and edit images with OpenAI's GPT Image 1.5 via CLI.

Community
Advanced