Help us improve
Share bugs, ideas, or general feedback.
From fal-ai-image
Generates and edits images using fal.ai nano-banana-pro model via Bash scripts. Supports text rendering in Cyrillic/CJK, infographics, complex compositions, and reference image editing.
npx claudepluginhub artwist-polyakov/polyakov-claude-skills --plugin fal-ai-imageHow this skill is triggered — by the user, by Claude, or both
Slash command
/fal-ai-image:fal-ai-imageThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Generate images via fal.ai nano-banana-pro (Gemini 3 Pro Image).
Generates AI images from text prompts, edits images, and composes from multiple references using Gemini models. Supports t2i, i2i, product mockups, and stickers.
Generates or edits images via Google Gemini from text prompts or up to 14 reference images for text-to-image and multi-image editing.
Generates and edits images using Google's Gemini Nano Banana Pro model (gemini-3-pro-image-preview) via bash API calls. Supports logos, product mockups, photo edits; requires GEMINI_API_KEY.
Share bugs, ideas, or general feedback.
Generate images via fal.ai nano-banana-pro (Gemini 3 Pro Image). Best for: infographics, text rendering, complex compositions.
Task with subagent_type: "Bash" and model: "haiku"upload.sh first to get URLs for edit.shUser gave reference images? → Edit mode (upload.sh → edit.sh)
User wants text-only gen? → Generate mode (generate.sh)
Multiple images needed? → Parallel Bash/haiku subagents
Scripts are POSIX sh compatible — work in cloud sandboxes (/bin/sh) and locally (bash).
No bashisms: [[ ]], ${BASH_SOURCE}, source etc. are NOT used.
Requires FAL_KEY in config/.env or environment.
Get key: https://fal.ai/dashboard/keys
Create images from text prompt only.
Script: scripts/generate.sh
Create images using reference images (up to 14).
Script: scripts/edit.sh
IMPORTANT: Run generation via Task tool with Haiku subagent to avoid blocking main context.
Clarify params (if not specified):
1:1, 16:9, 9:16, 4:3, etc.1K (default), 2K, 4KPropose save path based on project structure:
./images/, ./assets/, ./static/./images/infographic_coffee.png?"Show price & confirm:
Cost: $0.15/image (4K: $0.30)
Confirm? (yes/no)
Launch subagent:
Task tool:
- subagent_type: "Bash"
- model: "haiku"
- run_in_background: true
- prompt: Run generate.sh with params
Report result: Parse JSON output, show image URL, read saved file.
Get reference images:
upload.sh first to get URLsClarify prompt: What to do with references?
Show price & confirm: Same as generate
Launch subagent with edit.sh
Report result
sh scripts/generate.sh \
--prompt "infographic about coffee brewing" \
--aspect-ratio "9:16" \
--resolution "1K" \
--output-dir "./images" \
--filename "coffee_infographic"
| Param | Required | Default | Values |
|---|---|---|---|
--prompt | yes | - | text |
--aspect-ratio | no | 1:1 | 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16 |
--resolution | no | 1K | 1K, 2K, 4K |
--num-images | no | 1 | 1-4 |
--output-dir | no | - | path |
--filename | no | generated | base name |
--web-search | no | false | flag |
sh scripts/edit.sh \
--prompt "combine these into a collage" \
--image-urls "https://example.com/img1.png,https://example.com/img2.png" \
--aspect-ratio "16:9" \
--output-dir "./images" \
--filename "collage"
| Param | Required | Default | Values |
|---|---|---|---|
--prompt | yes | - | text |
--image-urls | yes | - | comma-separated URLs (max 14) |
--aspect-ratio | no | auto | auto, 1:1, 16:9, etc. |
--resolution | no | 1K | 1K, 2K, 4K |
--num-images | no | 1 | 1-4 variations |
--output-dir | no | - | path |
--filename | no | edited | base name |
# Get URL for local file
URL=$(sh scripts/upload.sh --file /path/to/image.png)
# Or get base64 data URI (for small files)
URI=$(sh scripts/upload.sh --file /path/to/image.png --base64)
For multiple images — launch several subagents in parallel:
Task 1: generate "cat in space" → cat_space.png
Task 2: generate "dog on moon" → dog_moon.png
Task 3: generate "bird in ocean" → bird_ocean.png
Each runs independently via Haiku, results collected when done.
Formula: price = num_images * (resolution == "4K" ? 0.30 : 0.15) + (web_search ? 0.015 : 0)