How do I install the Blink Image skill?

Run npx skillmds@latest add blink-new/blink-image 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 Blink Image safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Blink Image?

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 Blink Image free to use?

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

Who published Blink Image?

blink-new (@blink-new) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Blink Image by blink-new — Agent Skill

Name: Blink Image
Author: blink-new

Blink Image

Generate images from text prompts, or edit/transform existing images. Uses fal.ai models via Blink AI Gateway. Returns image URLs you can share, download, or use in further tasks. Use when asked to create, draw, generate, or edit any image.

blink-new bb263db 2.8 KB Updated Sep 17, 2026

File contents

Blink Image Generation

Generate images from text descriptions or edit existing images. All images are charged to your Blink workspace credits.

Generate an image from text

blink ai image "A serene mountain lake at golden hour, photorealistic"

Generate with a specific model

blink ai image "A futuristic city skyline" --model fal-ai/nano-banana-pro

Generate multiple images at once

blink ai image "A cozy coffee shop interior" --model fal-ai/nano-banana --n 4

Generate with a specific count and model

blink ai image "Product photo of a red sneaker on white background" --model fal-ai/nano-banana-pro --n 2

Note: Output format (jpeg/webp/png) is determined by the model, not a parameter.

Edit an existing image (image-to-image)

blink ai image-edit "Make this look like a watercolor painting" "https://example.com/photo.jpg"

Edit with high-quality model

blink ai image-edit "Add snow to this landscape" "https://example.com/landscape.jpg" --model fal-ai/nano-banana-pro/edit

Edit a LOCAL file (user uploaded a photo via Telegram/Discord/Slack)

When a user sends you a photo, OpenClaw saves it to disk. Use blink ai animate for local files (it handles upload automatically), or use blink ai image-edit with a URL:

# Animate/transform a local file directly (handles upload automatically)
blink ai animate "Make this a professional studio headshot with dark background" /data/agents/main/agent/photo.jpg

# Or use image-edit if you have a public URL
blink ai image-edit "Make this a professional studio headshot with dark background" "https://example.com/photo.jpg"

Find where OpenClaw saved an attachment

OpenClaw downloads attachments to the agent working directory. Check recent files:

ls -lt /data/agents/main/agent/ | head -10

Models available

Model	Best for
`fal-ai/nano-banana`	Default. Fast text→image.
`fal-ai/nano-banana-pro`	Higher quality text→image.
`fal-ai/nano-banana/edit`	Edit/transform existing images.
`fal-ai/nano-banana-pro/edit`	High-quality image editing.

Common use cases

"Generate a logo for my startup" → generate with clear prompt
"Create a product image on a white background" → generate
"Turn this photo into an oil painting" → image-edit with style prompt
"Make 4 variations of this design" → generate with --n 4
"Create a banner image for my Discord" → generate with aspect ratio guidance in prompt

---name: blink-imagedescription: Generate images from text prompts, or edit/transform existing images. Uses fal.ai models via Blink AI Gateway. Returns image URLs you can share, download, or use in further tasks. Use when asked to create, draw, generate, or edit any image.---# Blink Image GenerationGenerate images from text descriptions or edit existing images.All images are charged to your Blink workspace credits.## Generate an image from text```bashblink ai image "A serene mountain lake at golden hour, photorealistic"```## Generate with a specific model```bashblink ai image "A futuristic city skyline" --model fal-ai/nano-banana-pro```## Generate multiple images at once```bashblink ai image "A cozy coffee shop interior" --model fal-ai/nano-banana --n 4```## Generate with a specific count and model```bashblink ai image "Product photo of a red sneaker on white background" --model fal-ai/nano-banana-pro --n 2```> Note: Output format (jpeg/webp/png) is determined by the model, not a parameter.## Edit an existing image (image-to-image)```bashblink ai image-edit "Make this look like a watercolor painting" "https://example.com/photo.jpg"```## Edit with high-quality model```bashblink ai image-edit "Add snow to this landscape" "https://example.com/landscape.jpg" --model fal-ai/nano-banana-pro/edit```## Edit a LOCAL file (user uploaded a photo via Telegram/Discord/Slack)When a user sends you a photo, OpenClaw saves it to disk. Use `blink ai animate` for local files (it handles upload automatically), or use `blink ai image-edit` with a URL:```bash# Animate/transform a local file directly (handles upload automatically)blink ai animate "Make this a professional studio headshot with dark background" /data/agents/main/agent/photo.jpg# Or use image-edit if you have a public URLblink ai image-edit "Make this a professional studio headshot with dark background" "https://example.com/photo.jpg"```## Find where OpenClaw saved an attachmentOpenClaw downloads attachments to the agent working directory. Check recent files:```bashls -lt /data/agents/main/agent/ | head -10```## Models available| Model | Best for ||-------|---------|| `fal-ai/nano-banana` | Default. Fast text→image. || `fal-ai/nano-banana-pro` | Higher quality text→image. || `fal-ai/nano-banana/edit` | Edit/transform existing images. || `fal-ai/nano-banana-pro/edit` | High-quality image editing. |## Common use cases- "Generate a logo for my startup" → generate with clear prompt- "Create a product image on a white background" → generate- "Turn this photo into an oil painting" → image-edit with style prompt- "Make 4 variations of this design" → generate with --n 4- "Create a banner image for my Discord" → generate with aspect ratio guidance in prompt

Frequently asked questions