fal – OpenClaw Skill

---

name: fal description: Search, explore, and run fal.ai generative AI models (image generation, video, audio, 3D). Use when user wants to generate images, videos, or other media with AI models. allowed-tools: Bash(curl *), Bash(jq *), Bash(mkdir *), Read, Write argument-hint: "<command> [model_id] [--param value]"

fal.ai Model API Skill

Run 600+ generative AI models on fal.ai.

Arguments

• Command: $0 (search | schema | run | status | result | upload)
• Arg 1: $1 (model_id, search query, or file path)
• Arg 2+: $2, $3, etc. (additional parameters)
• All args: $ARGUMENTS

Session Output

Save generated files to session folder:

mkdir -p ~/.fal/sessions/${CLAUDE_SESSION_ID}

Downloaded images/videos go to: ~/.fal/sessions/${CLAUDE_SESSION_ID}/

---

Authentication

Requires FAL_KEY environment variable. If requests fail with 401, tell user:

Get an API key from https://fal.ai/dashboard/keys
Then: export FAL_KEY="your-key-here"

---

Command: $0

If $0 = "search"

Search for models matching $1:

curl -s "https://api.fal.ai/v1/models?q=$1&limit=15" \
  -H "Authorization: Key $FAL_KEY" | jq -r '.models[] | "• \(.endpoint_id) — \(.metadata.display_name) [\(.metadata.category)]"'

For category search, use:

curl -s "https://api.fal.ai/v1/models?category=$1&limit=15" \
  -H "Authorization: Key $FAL_KEY" | jq -r '.models[] | "• \(.endpoint_id) — \(.metadata.display_name)"'

Categories: text-to-image, image-to-video, text-to-video, image-to-3d, training, speech-to-text, text-to-speech

---

If $0 = "schema"

Get input schema for model $1:

curl -s "https://api.fal.ai/v1/models?endpoint_id=$1&expand=openapi-3.0" \
  -H "Authorization: Key $FAL_KEY" | jq '.models[0].openapi.components.schemas.Input.properties'

Show required vs optional fields to help user understand what inputs are needed.

---

If $0 = "run"

Run model $1 with parameters from remaining arguments.

Step 1: Parse parameters Extract --key value pairs from $ARGUMENTS after the model_id to build JSON payload.

Example: /fal run fal-ai/flux-2 --prompt "a cat" --image_size landscape_16_9 → Model: fal-ai/flux-2 → Payload: {"prompt": "a cat", "image_size": "landscape_16_9"}

Step 2: Submit to queue

curl -s -X POST "https://queue.fal.run/$1" \
  -H "Authorization: Key $FAL_KEY" \
  -H "Content-Type: application/json" \
  -d '<JSON_PAYLOAD>'

Step 3: Poll until complete

# Get request_id from response, then poll:
while true; do
  STATUS=$(curl -s "https://queue.fal.run/$1/requests/$REQUEST_ID/status" \
    -H "Authorization: Key $FAL_KEY" | jq -r '.status')
  echo "Status: $STATUS"
  if [ "$STATUS" = "COMPLETED" ]; then break; fi
  if [ "$STATUS" = "FAILED" ]; then echo "Job failed"; break; fi
  sleep 3
done

Step 4: Get result and save

# Fetch result
RESULT=$(curl -s "https://queue.fal.run/$1/requests/$REQUEST_ID" \
  -H "Authorization: Key $FAL_KEY")

# Create session output folder
mkdir -p ~/.fal/sessions/${CLAUDE_SESSION_ID}

# Download images/videos
# For images: jq -r '.images[0].url' and curl to download
# Save as: ~/.fal/sessions/${CLAUDE_SESSION_ID}/<timestamp>_<model>.png

---

If $0 = "status"

Check status of request $2 for model $1:

curl -s "https://queue.fal.run/$1/requests/$2/status?logs=1" \
  -H "Authorization: Key $FAL_KEY" | jq '{status: .status, queue_position: .queue_position, logs: .logs}'

---

If $0 = "result"

Get result of completed request $2 for model $1:

curl -s "https://queue.fal.run/$1/requests/$2" \
  -H "Authorization: Key $FAL_KEY" | jq '.'

---

If $0 = "upload"

Upload file $1 to fal CDN:

curl -s -X POST "https://fal.run/fal-ai/storage/upload" \
  -H "Authorization: Key $FAL_KEY" \
  -F "file=@$1"

Returns URL to use in model requests.

---

Quick Reference

Popular models:

• fal-ai/flux-2 — Fast text-to-image
• fal-ai/flux-2-pro — High quality text-to-image
• fal-ai/kling-video/v2/image-to-video — Image to video
• fal-ai/minimax/video-01/image-to-video — Image to video
• fal-ai/whisper — Speech to text

Common parameters for text-to-image:

• --prompt "description" — What to generate
• --image_size landscape_16_9 — Aspect ratio (square, portrait_4_3, landscape_16_9)
• --num_images 1 — Number of images

Example invocations:

• /fal search video — Find video models
• /fal schema fal-ai/flux-2 — See input options
• /fal run fal-ai/flux-2 --prompt "a sunset over mountains"
• /fal status fal-ai/flux-2 abc-123
• /fal upload ./photo.png

fal-skill

A Claude Code skill for running fal.ai generative AI models - image generation, video, audio, 3D, and more.

Installation

git clone https://github.com/fal-ai/fal-skill ~/.claude/skills/fal

Setup

1. Get an API key from fal.ai/dashboard/keys

2. Add to your shell config (~/.zshrc or ~/.bashrc):

   export FAL_KEY="your-key-here"
   

3. Reload your shell:

   source ~/.zshrc
   

Usage

Commands

Command	Description
/fal search <query>	Search 600+ models
/fal schema <model_id>	Get input/output schema
/fal run <model_id> --param value	Run a model
/fal status <model_id> <request_id>	Check job status
/fal result <model_id> <request_id>	Get job result
/fal upload <file>	Upload file to fal CDN

Examples

# Search for models
/fal search video

# See what inputs a model accepts
/fal schema fal-ai/flux-2

# Generate an image
/fal run fal-ai/flux-2 --prompt "a cat in space"

Or just ask Claude naturally:

"Generate an image of a sunset over mountains"

"Turn this photo into a video"

"What image models are available?"

Popular Models

Model	Category
fal-ai/flux-2	Text to image
fal-ai/flux-2-pro	Text to image (high quality)
fal-ai/kling-video/v2/image-to-video	Image to video
fal-ai/minimax/video-01/image-to-video	Image to video
fal-ai/whisper	Speech to text

See models-reference.md for a full list.

How It Works

This skill teaches Claude how to interact with fal.ai APIs using curl. When you ask Claude to generate media, it:

1. Searches for appropriate models (if needed)
2. Fetches the model schema to understand required inputs
3. Submits the job to fal's queue
4. Polls for completion
5. Downloads results to your machine

Links

• fal.ai Models - Browse all available models
• fal.ai Docs - API documentation
• Claude Code Skills - Learn about skills

License

MIT