Official skills from Black Forest Labs for FLUX image generation models. These skills provide prompting guidelines and API integration patterns following the agentskills.io specification.

npx skills add black-forest-labs/skills

Or install individual skills:

# FLUX best practices only
npx skills add black-forest-labs/skills --skill flux-best-practices

# API integration only
npx skills add black-forest-labs/skills --skill bfl-api

You can also add this as a plugin marketplace in Claude Code:

/plugin marketplace add black-forest-labs/skills
/plugin install flux-best-practices@black-forest-labs

Comprehensive guide for all FLUX models including:

• Core Principles - Universal prompting best practices
• Model-Specific Guides - FLUX.2 ([klein], [max], [pro], [flex], [dev]) and FLUX.1
• T2I Prompting - Text-to-image patterns and techniques
• I2I Prompting - Image-to-image editing with FLUX.2 reference images
• JSON Structured Prompting - Complex scene composition
• Hex Color Prompting - Precise color specification (#RRGGBB)
• Typography - Text rendering and font styles
• Multi-Reference Editing - Using multiple reference images
• Negative Prompt Alternatives - Positive replacements (FLUX doesn't support negatives)
• Model Selection Guide - Choosing the right model for your use case

API integration guide covering:

• Endpoints - Complete endpoint documentation for all FLUX.2 and FLUX.1 models
• Polling Patterns - Async polling with exponential backoff
• Rate Limiting - Handling 24 concurrent requests
• Error Handling - Error codes and recovery strategies
• Webhook Integration - Production webhook setup and verification
• Code Examples - Python and TypeScript clients

All FLUX.2 models support both text-to-image and image editing natively—no need for separate models. FLUX.1 models are also available.

[Subject] + [Action] + [Style] + [Context] + [Lighting] + [Technical]

1. NO negative prompts - FLUX doesn't support them; describe what you want
2. Be specific - More detail yields better results
3. Use natural language - Prose style works best
4. Specify lighting - Has the biggest impact on quality
5. Quote text - Use "quoted text" for typography
6. Hex colors - Use #RRGGBB with color names

import requests
import time

API_KEY = "your-api-key"
BASE_URL = "https://api.bfl.ai"

# Submit request
response = requests.post(
    f"{BASE_URL}/v1/flux-2-pro",
    headers={"x-key": API_KEY},
    json={"prompt": "A serene mountain landscape"}
)
polling_url = response.json()["polling_url"]

# Poll for result
while True:
    result = requests.get(polling_url, headers={"x-key": API_KEY})
    data = result.json()
    if data["status"] == "Ready":
        print(f"Result: {data['result']}")  # Expires in 10 min
        break
    time.sleep(2)

• BFL Documentation
• API Reference
• Prompting Guides

MIT

Black Forest Labs