Skip to content
Back to search

black-forest-labs

MarketplaceWorks with Claude CodeGitHub

Official agent skills from Black Forest Labs for FLUX image and video generation — prompting guides and API integration patterns for Claude Code, Codex, and any agentskills.io-compatible agent.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/black-forest-labs--skills/badge.svg)](/listings/black-forest-labs--skills)

HTML

<a href="/listings/black-forest-labs--skills"><img src="/api/listings/black-forest-labs--skills/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Everything at once:

npx skills add black-forest-labs/skills

Or start with one skill and add specialists as you need them:

# FLUX 3 video: the router pulls in the right specialist per request
npx skills add black-forest-labs/skills --skill flux-3-video

# Image prompting for every FLUX model
npx skills add black-forest-labs/skills --skill flux-image-best-practices

# BFL API integration (polling, webhooks, error handling)
npx skills add black-forest-labs/skills --skill bfl-api

<details> <summary>Install individual FLUX 3 specialists</summary>

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-ads

</details>

Claude Code plugin

/plugin marketplace add black-forest-labs/skills
# Image skills
/plugin install flux-image-best-practices@black-forest-labs
# FLUX 3 video suite
/plugin install flux-3-video@black-forest-labs

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 46 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. {
  2. "$schema": "https://anthropic.com/claude-code/marketplace.schema.json",
  3. "name": "black-forest-labs",
  4. "owner": {
  5. "name": "Black Forest Labs",
  6. "url": "https://blackforestlabs.ai"
  7. },
  8. "metadata": {
  9. "description": "Official skills from Black Forest Labs for FLUX image and video generation - prompting guidelines and API integration patterns",
  10. "version": "1.1.0"
  11. },
  12. "plugins": [
  13. {
  14. "name": "flux-image-best-practices",
  15. "description": "BFL skills for image generation including best practices for prompting, T2I, I2I, structured JSON, and API integration patterns",
  16. "source": "./",
  17. "strict": false,
  18. "skills": [
  19. "./skills/bfl-api",
  20. "./skills/flux-image-best-practices"
  21. ],
  22. "author": {
  23. "name": "Kerem Nalbant"
  24. }
  25. },
  26. {
  27. "name": "flux-3-video",
  28. "description": "BFL skills for FLUX 3 video generation including prompt readiness checks, cinematic shot craft for text-to-video, keyframes and video continuation, synchronized audio and dialogue, period and archival format looks, request construction with async polling and downloads, and finished product ads with voiceover and deterministic assembly",
  29. "source": "./",
  30. "strict": false,
  31. "skills": [
  32. "./skills/flux-3-video",
  33. "./skills/flux-3-prompt-doctor",
  34. "./skills/flux-3-cinematic-inserts",
  35. "./skills/flux-3-keyframes-continuation",
  36. "./skills/flux-3-audio-dialogue",
  37. "./skills/flux-3-generate",
  38. "./skills/flux-3-archival-formats",
  39. "./skills/flux-3-product-ads"
  40. ],
  41. "author": {
  42. "name": "Black Forest Labs"
  43. }
  44. }
  45. ]
  46. }
README.md · README · 189 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <h1 align="center">BFL Skills</h1>
  2. <h4 align="center">Official <a href="https://blackforestlabs.ai">Black Forest Labs</a> skills for FLUX image and video generation.</h4>
  3. <p align="center">
  4. <a href="#installation">Installation</a> •
  5. <a href="#whats-inside">What's inside</a> •
  6. <a href="#flux-3-quick-reference">FLUX 3 reference</a> •
  7. <a href="#flux-image-quick-reference">Image reference</a> •
  8. <a href="#documentation">Documentation</a> •
  9. <a href="#license">License</a>
  10. </p>
  11. ---
  12. Each skill teaches your coding agent how to prompt a FLUX model well and drive the [BFL API](https://docs.bfl.ai) correctly, following the [agentskills.io](https://agentskills.io) specification.
  13. Every clip in the reel below came out of FLUX 3 in one request, sound and all: a night-market conversation in Thai, a honey macro, a paper boat in stop motion, sparklers on Super 8, rain in a noir alley.
  14. <p align="center">
  15. <a href="https://cdn.sanity.io/files/2gpum2i6/production/1e6e576489baa4e734bbae4cb62dcd383151e030.mp4"><img src="https://cdn.sanity.io/images/2gpum2i6/production/c7ef05daf120891bdc96e5e2a9c2b9947d206d65-1920x1088.jpg?w=600" alt="FLUX 3 style reel: five styles, one model, audio on every shot"></a>
  16. </p>
  17. ## Installation
  18. Everything at once:
  19. ```bash
  20. npx skills add black-forest-labs/skills
  21. ```
  22. Or start with one skill and add specialists as you need them:
  23. ```bash
  24. # FLUX 3 video: the router pulls in the right specialist per request
  25. npx skills add black-forest-labs/skills --skill flux-3-video
  26. # Image prompting for every FLUX model
  27. npx skills add black-forest-labs/skills --skill flux-image-best-practices
  28. # BFL API integration (polling, webhooks, error handling)
  29. npx skills add black-forest-labs/skills --skill bfl-api
  30. ```
  31. <details>
  32. <summary>Install individual FLUX 3 specialists</summary>
  33. ```bash
  34. npx skills add black-forest-labs/skills --skill flux-3-prompt-doctor
  35. npx skills add black-forest-labs/skills --skill flux-3-cinematic-inserts
  36. npx skills add black-forest-labs/skills --skill flux-3-keyframes-continuation
  37. npx skills add black-forest-labs/skills --skill flux-3-audio-dialogue
  38. npx skills add black-forest-labs/skills --skill flux-3-generate
  39. npx skills add black-forest-labs/skills --skill flux-3-archival-formats
  40. npx skills add black-forest-labs/skills --skill flux-3-product-ads
  41. ```
  42. </details>
  43. ### Claude Code plugin
  44. ```bash
  45. /plugin marketplace add black-forest-labs/skills
  46. # Image skills
  47. /plugin install flux-image-best-practices@black-forest-labs
  48. # FLUX 3 video suite
  49. /plugin install flux-3-video@black-forest-labs
  50. ```
  51. ## What's inside
  52. | Skill | What it teaches your agent |
  53. | --- | --- |
  54. | `flux-image-best-practices` | Prompting for every FLUX model: T2I and I2I patterns, JSON structured prompts, hex colors, typography, multi-reference editing, model selection |
  55. | `bfl-api` | The BFL API end to end: endpoints, async polling, rate limits, error recovery, webhooks, Python and TypeScript clients |
  56. | `flux-3-video` (suite, below) | Video generation with native audio, from a single shot to a finished product ad |
  57. ### The FLUX 3 video suite
  58. Eight skills that hand off to each other. Install the router alone and it names the specialist a request needs; install only the specialists you use.
  59. | Skill | Use it to |
  60. | --- | --- |
  61. | `flux-3-video` | Route a general FLUX 3 request to the right specialist |
  62. | `flux-3-prompt-doctor` | Get a readiness verdict (`READY` / `NEEDS INFO` / `REVISE`) before spending a generation |
  63. | `flux-3-cinematic-inserts` | Write text-to-video prompts with real shot craft, from a field guide of proven concepts |
  64. | `flux-3-keyframes-continuation` | Build from your own images (`i2v`) or continue a clip (`v2v`) |
  65. | `flux-3-audio-dialogue` | Direct speech, voiceover, ambience, effects, and music |
  66. | `flux-3-generate` | Construct requests, poll, download, and validate the result |
  67. | `flux-3-archival-formats` | Give a shot a period look by naming the format: Super 8, VHS, 16mm, Hi8 |
  68. | `flux-3-product-ads` | Assemble a finished product ad: voiceover, action-to-word sync, QC gates |
  69. ## FLUX 3 quick reference
  70. ### Modes
  71. Every request names its `mode` and carries the matching media field:
  72. | `mode` | Media field | What the model does |
  73. | --- | --- | --- |
  74. | `t2v` | none | Generates the clip from your text alone |
  75. | `i2v` | `keyframes` | Puts your images on screen, pixel-exact, as pinned frames |
  76. | `v2v` | `start_video` | Continues from the final frames of your clip |
  77. | `draft_enhance` | `draft_cache` | Replays a chosen draft at full quality |
  78. ### Draft, then commit
  79. 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 with the same prompt and settings, so the final render is the shot you approved rather than a fresh attempt.
  80. Judge a draft on event legibility, composition, and continuity. Softness and coarse texture are artifacts of low-step rendering and resolve at full quality.
  81. ## FLUX image quick reference
  82. Four FLUX.2 [max] generations, one prompt each, no retries. Clockwise from top left: exact quoted typography, hex-specified colors (`#2F5D50`, `#E8A13D`, `#F2EDE4`), flat-vector illustration, photoreal macro detail.
  83. <table>
  84. <tr>
  85. <td><img src="https://cdn.sanity.io/images/2gpum2i6/production/00787272007afd69df1dd1098ea898639e8d7171-1920x1088.jpg?w=800" alt="Letterpress storefront with exact quoted lettering: INK & QUARREL, EST. 1962 - FINE PRINTING"></td>
  86. <td><img src="https://cdn.sanity.io/images/2gpum2i6/production/4a4555f9ff1a03fea72d57b51ff0173554060a39-1920x1088.jpg?w=800" alt="Ceramic pour-over set with hex-specified green, amber, and off-white palette"></td>
  87. </tr>
  88. <tr>
  89. <td><img src="https://cdn.sanity.io/images/2gpum2i6/production/a642526cd6e434a09a014efafeb5e9e515057807-1920x1088.jpg?w=800" alt="Photoreal beekeeper close-up with bees in flight at golden hour"></td>
  90. <td><img src="https://cdn.sanity.io/images/2gpum2i6/production/921b94de38e77ef516f71442c1ccd245db9fb6a9-1920x1088.jpg?w=800" alt="Isometric flat-vector cutaway of a lighthouse interior"></td>
  91. </tr>
  92. </table>
  93. ### Model selection
  94. | Model | Best for | Pricing |
  95. | -------------- | --------------------------------- | --------------------- |
  96. | FLUX.2 [klein] | Fastest generation, real-time | from $0.014/image |
  97. | FLUX.2 [pro] | Production balanced | from $0.03/MP |
  98. | FLUX.2 [flex] | Typography/text | from $0.06/MP |
  99. | FLUX.2 [max] | Highest quality, grounding search | from $0.07/MP |
  100. | FLUX.2 [dev] | Local development | Free (non-commercial) |
  101. All FLUX.2 models handle both text-to-image and image editing natively. FLUX.1 models are also available.
  102. ### Prompt structure
  103. ```
  104. [Subject] + [Action] + [Style] + [Context] + [Lighting] + [Technical]
  105. ```
  106. ### Core rules
  107. 1. **No negative prompts.** FLUX doesn't support them; describe what you want instead.
  108. 2. **Be specific.** More detail yields better results.
  109. 3. **Use natural language.** Prose beats keyword lists.
  110. 4. **Specify lighting.** It has the biggest impact on quality.
  111. 5. **Quote text.** Use "quoted text" for typography.
  112. 6. **Hex colors.** Use #RRGGBB alongside color names.
  113. ### API quick start
  114. ```python
  115. import requests
  116. import time
  117. API_KEY = "your-api-key"
  118. BASE_URL = "https://api.bfl.ai"
  119. # Submit request
  120. response = requests.post(
  121. f"{BASE_URL}/v1/flux-2-pro",
  122. headers={"x-key": API_KEY},
  123. json={"prompt": "A serene mountain landscape"}
  124. )
  125. polling_url = response.json()["polling_url"]
  126. # Poll for result
  127. while True:
  128. result = requests.get(polling_url, headers={"x-key": API_KEY})
  129. data = result.json()
  130. if data["status"] == "Ready":
  131. print(f"Result: {data['result']}") # Expires in 10 min
  132. break
  133. time.sleep(2)
  134. ```
  135. ## Documentation
  136. - [BFL Documentation](https://docs.bfl.ai)
  137. - [API Reference](https://docs.bfl.ai/api)
  138. - [Prompting Guides](https://docs.bfl.ai/guides)
  139. ## License
  140. MIT

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.