Official skills from Black Forest Labs for FLUX image and video generation models. These skills provide prompting guidelines and API integration patterns following the agentskills.io specification.
npx skills add black-forest-labs/skillsOr install individual skills:
# FLUX best practices only
npx skills add black-forest-labs/skills --skill flux-image-best-practices
# API integration only
npx skills add black-forest-labs/skills --skill bfl-api
# FLUX 3 video router
npx skills add black-forest-labs/skills --skill flux-3-video
# FLUX 3 specialist skills
npx skills add black-forest-labs/skills --skill flux-3-prompt-doctor
npx skills add black-forest-labs/skills --skill flux-3-cinematic-inserts
npx skills add black-forest-labs/skills --skill flux-3-keyframes-continuation
npx skills add black-forest-labs/skills --skill flux-3-audio-dialogue
npx skills add black-forest-labs/skills --skill flux-3-generate
npx skills add black-forest-labs/skills --skill flux-3-archival-formats
npx skills add black-forest-labs/skills --skill flux-3-product-adsYou can also add this as a plugin marketplace in Claude Code:
/plugin marketplace add black-forest-labs/skills
# Image skills: bfl-api and flux-image-best-practices
/plugin install flux-image-best-practices@black-forest-labs
# FLUX 3 video skills
/plugin install flux-3-video@black-forest-labsComprehensive 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
Eight skills that hand off to each other; install only what you need:
flux-3-video- thin router across the seven specialistsflux-3-prompt-doctor- readiness verdicts (READY/NEEDS INFO/REVISE) before generationflux-3-cinematic-inserts- shot craft for text-only generation, with a field guide of proven conceptsflux-3-keyframes-continuation- keyframes (i2v) and video continuation (v2v)flux-3-audio-dialogue- dialogue, voiceover, ambience, effects, musicflux-3-generate- request construction, polling, drafts, downloads, validationflux-3-archival-formats- period looks by naming a recording format (Super 8, VHS, 16mm, Hi8)flux-3-product-ads- finished product ads: voiceover, action-to-word sync, deterministic assembly, QC gates
Built with flux-3-product-ads: a 13.6s spot for an invented pen brand, three
plates, generated voiceover, action slipped to land on the word that names it.
Click to play with sound.
| Request | Skill |
|---|---|
| Diagnose or repair a brief | flux-3-prompt-doctor |
| Write a text-to-video prompt for a new shot | flux-3-cinematic-inserts |
| Build from supplied images or video | flux-3-keyframes-continuation |
| Direct speech, ambience, effects, or music | flux-3-audio-dialogue |
| Submit, poll, enhance, download, or technically validate | flux-3-generate |
| Build a finished product ad with voiceover | flux-3-product-ads |
| Give a shot a period or archival look | flux-3-archival-formats |
| Route a general FLUX 3 video request | flux-3-video |
Every request names its mode and carries the matching media field:
mode |
Media field | What the model does |
|---|---|---|
t2v |
none | Generates the clip from your text alone |
i2v |
keyframes |
Puts your images on screen, pixel-exact, as pinned frames |
v2v |
start_video |
Continues from the final frames of your clip |
draft_enhance |
draft_cache |
Replays a chosen draft at full quality |
A full render takes several minutes. Adding draft: true returns a fast, low-step preview plus a draft_cache, so you can settle concept and composition cheaply before paying for finish. draft_enhance then replays that cache at full quality (same prompt and settings), so the final render is the shot you approved rather than a fresh attempt.
Judge a draft on event legibility, composition, and continuity. Softness and coarse texture are artifacts of low-step rendering and resolve at full quality.
See docs.bfl.ai for current fields, limits, and endpoints.
| Model | Best For | Pricing |
|---|---|---|
| FLUX.2 [klein] | Fastest generation, real-time | from $0.014/image |
| FLUX.2 [pro] | Production balanced | from $0.03/MP |
| FLUX.2 [flex] | Typography/text | from $0.06/MP |
| FLUX.2 [max] | Highest quality, grounding search | from $0.07/MP |
| FLUX.2 [dev] | Local development | Free (non-commercial) |
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]
- NO negative prompts - FLUX doesn't support them; describe what you want
- Be specific - More detail yields better results
- Use natural language - Prose style works best
- Specify lighting - Has the biggest impact on quality
- Quote text - Use "quoted text" for typography
- 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)MIT