How do I install the Ivx Cf Image Generation skill?

Run npx skillmds@latest add intelli-verse-x/ivx-cf-image-generation in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Ivx Cf Image Generation safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Ivx Cf Image Generation?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Ivx Cf Image Generation free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Ivx Cf Image Generation?

intelli-verse-x (@intelli-verse-x) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Ivx Cf Image Generation by intelli-verse-x

Name: Ivx Cf Image Generation
Rating: 4.2 (5 reviews)
Author: intelli-verse-x

File contents

Image Generation Skill

Purpose

Generate high-quality images through structured, reproducible pipelines.

Pipeline Stages

Concept: Purpose, style, composition, subject
Prompt Engineering: Detailed text prompt + negative prompt
Structure Control: ControlNet, depth, pose, canny
Generation: Base model + sampler + steps + CFG
Selection: Generate multiple, pick best
Post-Processing: Upscale, inpaint, color correct
Export: Format, resolution, color space

Generation Config

Parameter	Typical Value	Notes
Model	FLUX.1-dev / SDXL	Quality vs speed tradeoff
Resolution	1024x1024	Native resolution for most models
Sampler	DPM++ 2M Karras	Good balance quality/speed
Steps	20-50	More steps = better quality, slower
CFG	7-12	Higher = stricter adherence
Seed	Random / fixed	Fixed for reproducibility

Control Methods

Method	Use Case	Strength
ControlNet (canny)	Edge structure	0.8-1.0
ControlNet (depth)	Spatial depth	0.8-1.0
ControlNet (pose)	Human pose	0.8-1.2
IP-Adapter	Style transfer	0.6-0.8
Reference-only	Character consistency	0.5-0.7

CF-Specific

# 1. Generate with style consistency
image = await generate_image(
    prompt="A cinematic shot of...",
    negative_prompt="blurry, low quality, watermark",
    width=1024, height=1024,
    seed=42,  # Fixed for reproducibility
    controlnet={"type": "canny", "image": reference, "strength": 0.9},
)

# 2. Upscale
image = await upscale(image, scale=4, model="Real-ESRGAN")

# 3. Save to output_path
output_path.write_image(image, filename="scene_01.png")

Quality Checklist

Prompt adherence (subject matches request)
No artifacts (extra limbs, distorted faces)
Style consistency (matches references)
Resolution meets target
Color accuracy (if brand colors specified)
No unintended text/watermarks

Batch Processing

Group by style/model for efficiency
Use same seed + prompt variations for consistency
Parallel generation (GPU memory permitting)
Post-process batch with same settings

---name: ivx-cf-image-generationdescription: Design and execute image generation pipelines using diffusion models. Use when generating images, optimizing image quality, or building image production workflows.---# Image Generation Skill## PurposeGenerate high-quality images through structured, reproducible pipelines.## Pipeline Stages1. **Concept**: Purpose, style, composition, subject2. **Prompt Engineering**: Detailed text prompt + negative prompt3. **Structure Control**: ControlNet, depth, pose, canny4. **Generation**: Base model + sampler + steps + CFG5. **Selection**: Generate multiple, pick best6. **Post-Processing**: Upscale, inpaint, color correct7. **Export**: Format, resolution, color space## Generation Config| Parameter | Typical Value | Notes ||-----------|--------------|-------|| Model | FLUX.1-dev / SDXL | Quality vs speed tradeoff || Resolution | 1024x1024 | Native resolution for most models || Sampler | DPM++ 2M Karras | Good balance quality/speed || Steps | 20-50 | More steps = better quality, slower || CFG | 7-12 | Higher = stricter adherence || Seed | Random / fixed | Fixed for reproducibility |## Control Methods| Method | Use Case | Strength ||--------|----------|----------|| ControlNet (canny) | Edge structure | 0.8-1.0 || ControlNet (depth) | Spatial depth | 0.8-1.0 || ControlNet (pose) | Human pose | 0.8-1.2 || IP-Adapter | Style transfer | 0.6-0.8 || Reference-only | Character consistency | 0.5-0.7 |## CF-Specific```python# 1. Generate with style consistencyimage = await generate_image( prompt="A cinematic shot of...", negative_prompt="blurry, low quality, watermark", width=1024, height=1024, seed=42, # Fixed for reproducibility controlnet={"type": "canny", "image": reference, "strength": 0.9},)# 2. Upscaleimage = await upscale(image, scale=4, model="Real-ESRGAN")# 3. Save to output_pathoutput_path.write_image(image, filename="scene_01.png")```## Quality Checklist- [ ] Prompt adherence (subject matches request)- [ ] No artifacts (extra limbs, distorted faces)- [ ] Style consistency (matches references)- [ ] Resolution meets target- [ ] Color accuracy (if brand colors specified)- [ ] No unintended text/watermarks## Batch Processing- Group by style/model for efficiency- Use same seed + prompt variations for consistency- Parallel generation (GPU memory permitting)- Post-process batch with same settings

Frequently asked questions

Ivx Cf Image Generation