스킬 flux-2-klein

🎨

flux-2-klein

Name: flux-2-klein
Author: runcomfy-com

낮은 위험 ⚙️ 외부 명령어🌐 네트워크 접근📁 파일 시스템 액세스

Generate Images with Flux 2 Klein on RunComfy

또한 다음에서 사용할 수 있습니다: doany-ai,agentspace-so

Image generation models can be difficult to prompt optimally. This skill provides Flux 2 Klein-specific patterns including step-count strategy, subject-first grammar, and model routing guidance for consistent professional results.

지원: Claude Codex Code(CC)

📊 71 적절함

스킬 ZIP 다운로드

Claude에서 업로드

설정 → 기능 → 스킬 → 스킬 업로드로 이동

토글을 켜고 사용 시작

테스트해 보기

"flux-2-klein" 사용 중입니다. A sleek wireless headphone product shot on matte white surface, soft studio lighting, 50mm lens, shallow depth of field, minimal background, professional product photography

예상 결과:

A high-quality image of modern wireless headphones displayed cleanly on a white background with soft, professional studio lighting and subtle depth-of-field effect

"flux-2-klein" 사용 중입니다. Same composition as reference, but change the background from outdoor park to indoor coffee shop with warm ambient lighting

예상 결과:

The original subject and composition preserved, now set inside a cozy coffee shop with warm lighting, wooden furniture visible in the background, maintaining the original framing and subject pose

"flux-2-klein" 사용 중입니다. A vintage botanical illustration of lavender in a copper vase, hand-drawn style, muted earth tones, textured paper background

예상 결과:

An elegant vintage-style botanical rendering of purple lavender arranged in an antique copper vase, with delicate hand-drawn linework and a textured cream paper backdrop

보안 감사

낮은 위험

v1 • 6/1/2026

After evaluating the 83 static findings, 81 are false positives from markdown documentation being misidentified as executable code. The remaining findings reflect legitimate documented behavior: invoking the RunComfy CLI, network calls to model-api.runcomfy.net, and filesystem access for output files. Security documentation in SKILL.md (lines 217-223) confirms shell injection prevention, token file permissions (0600), and download URL whitelisting.

스캔된 파일

224

분석된 줄 수

발견 사항

총 감사 수

중간 위험 문제 (1)

SKILL.md:82-86

CLI Command Invocation

Skill invokes the runcomfy CLI to execute model API calls. Arguments include user-provided prompts passed via --input JSON. CLI documented to prevent shell injection by transmitting JSON body directly rather than shell-expanding.

낮은 위험 문제 (1)

SKILL.md:208-209

URL Pattern Matching for Downloads

CLI downloads files from URLs matching .runcomfy.net or .runcomfy.com hosts. Pattern uses simple string suffix matching.

위험 요인

⚙️ 외부 명령어 (4)

SKILL.md:27-29 SKILL.md:82-86 SKILL.md:90-99 SKILL.md:103-107

🌐 네트워크 접근 (3)

SKILL.md:17 SKILL.md:23 SKILL.md:207

📁 파일 시스템 액세스 (1)

SKILL.md:208-209

감사자: claude

품질 점수

아키텍처

100

유지보수성

콘텐츠

커뮤니티

보안

사양 준수

만들 수 있는 것

Rapid Product Visualization

Iterate quickly on product designs with sub-second feedback from the 4B variant. Generate multiple concept variations in seconds for client reviews.

Brand-Consistent Asset Generation

Use multi-reference alignment to maintain consistent brand aesthetics across marketing materials, social posts, and campaign visuals.

Live Art Direction Sessions

Engage in real-time creative iteration with clients, adjusting prompts and styles instantly to explore creative directions together.

이 프롬프트를 사용해 보세요

Basic Product Photo

Generate a [product type] on a [surface/background], with [lighting style], [camera settings], professional e-commerce ready

Subject-First Complex Scene

A [subject] [action/pose] [setting details], [style/medium], [lighting], [camera/technical specs], [quality modifiers]

Reference-Based Edit

Same [composition/lighting/subject] as reference, but [specific changes]. Keep [unchanged elements] exactly as in reference.

Multi-Reference Brand Style

Style consistent with reference images: [describe aesthetic elements present in all refs]. Subject: [new subject description]

모범 사례

Front-load the subject in prompts: state what the subject is before describing actions or settings
Use step counts strategically: 4-8 steps on 4B for rapid iteration, ~25 steps on 9B for final output
Keep reference images stylistically consistent when using multi-reference workflows

피하기

Do not use conflicting adjectives like 'minimalist AND ornate' as they cancel each other out
Do not exceed approximately 512 tokens in prompts as quality degrades without graceful truncation
Do not request 4K resolution or ultra-wide aspect ratios beyond 16:9 as the model cannot produce them

자주 묻는 질문

Which variant should I use: 4B or 9B?

Use the 4B variant for rapid iteration and concepting where sub-second response time matters. Use the 9B variant for final polish and high-quality output where you need maximum detail.

How do I maintain brand consistency across images?

Use multi-reference alignment with 2-4 reference images that share the same visual aesthetic. Keep all references in a consistent style (photorealistic, illustration, etc.) to guide the model effectively.

Why does my prompt exceed token limits?

Flux 2 Klein truncates at approximately 512 tokens rather than truncating gracefully. Remove redundant adjectives, repeated concepts, and quality modifiers once you have established the style.

When should I route to a different model instead?

Route to Flux 2 Pro for maximum prompt adherence and detail. Use Seedream 5 for 2K-4K resolution. Choose GPT Image 2 for embedded text, logos, or multilingual signage.

Can I edit existing images?

Yes, use reference images for guided composition. Pass up to 4 reference images and describe what should stay the same versus what should change using conditional edit patterns.

Why does text in images come out blurry?

Add crisp typography keywords and use approximately 25 steps on the 9B variant. For heavy text rendering or multilingual content, route to GPT Image 2 which handles embedded text better.

개발자 세부 정보

작성자

runcomfy-com

라이선스

MIT

리포지토리

https://github.com/runcomfy-com/skills/tree/main/flux-2-klein/

참조

main

파일 구조

📄 SKILL.md