Index of every page on this site: llms.txt
All skills
openmontageinc avatar

/flux-best-practices

@2e003e6
by openmontageincopenmontageinc/openmontage82k stars
11,135
Run it yourself

Comprehensive guide for BFL FLUX image generation models. Covers prompting, T2I, I2I, structured JSON, hex colors, typography, multi-reference editing, and model-specific best practices for FLUX.2 and FLUX.1 families.

Use this Skill: https://skilld.dev/gh/openmontageinc/openmontage/flux-best-practices

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈60 tokens for metadata: the name and description. ≈974 when used: this file. ≈30k more on demand in 13 files.

Description uses 3.0% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈239 characters, or 3.0%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

FLUX Best Practices

Use this skill when generating prompts for any BFL FLUX model to ensure optimal image quality and accurate prompt interpretation.

Extended reference: AGENTS.md in this directory is the long-form upstream FLUX guide (vendored from Black Forest Labs). It is supplementary reference material scoped to this skill only — SKILL.md is the loadable entry point and the authority. It does not override or extend the repository-root AGENTS.md / AGENT_GUIDE.md.

When to Use

  • Creating prompts for FLUX.2 or FLUX.1 models
  • Text-to-image (T2I) generation
  • Image-to-image (I2I) editing with FLUX.2 models
  • Structured scene generation with JSON
  • Typography and text rendering
  • Multi-reference style transfer
  • Color-accurate brand generations

Quick Reference

Prompt Structure Formula

[Subject] + [Action/Pose] + [Style/Medium] + [Context/Setting] + [Lighting] + [Camera/Technical]

Model Selection

Use Case Recommended Model Notes
Fastest generation FLUX.2 [klein] 4B or 9B, sub-second
Highest quality FLUX.2 [max] Best detail, grounding search
Production balanced FLUX.2 [pro] Quality + speed
Typography/text FLUX.2 [flex] Best text rendering
Local/development FLUX.2 [dev] Open weights
Image editing FLUX.2 [pro/max] Pass image URL directly to input_image
Inpainting FLUX.1 Fill Object removal/completion
Context editing FLUX.1 Kontext Older model, prefer FLUX.2

Critical Rules

  1. NO negative prompts - FLUX does not support negative prompts; describe what you want
  2. Be specific - Vague prompts produce mediocre results
  3. Use natural language - Prose/narrative style works best
  4. Specify lighting - Lighting has the biggest impact on quality
  5. Quote text - Use "quoted text" for typography rendering
  6. Hex colors - Use #RRGGBB format with color description

Related

For API integration (endpoints, polling, webhooks), see the bfl-api skill.

Rules Reference

Read individual rule files for detailed guidance:

Example Prompt

A weathered fisherman in his 70s with deep wrinkles and a salt-and-pepper beard,
wearing a navy cable-knit sweater, standing at the helm of his wooden boat.
Golden hour sunlight from the left creates dramatic rim lighting on his profile.
Shot on Hasselblad with 85mm lens at f/2.8, shallow depth of field with harbor
lights creating soft bokeh in the background. Kodak Portra 400 color science.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts12d3 checks · Risk SAFE
  • Gen Agent Trust Hub12d

    The flux-best-practices skill is a documentation-only guide for optimizing image generation using Black Forest Labs FLUX models. It contains no executable code or risky configurations and adheres to standard prompting best practices.

  • Socket12d

    No alerts

  • Snyk12d

    Risk: LOW · No issues

Signed by skilld at 2e003e6. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 4 days ago.

Activeupdated 5 months ago
Other metadata
metadata
{
  "author": "Black Forest Labs",
  "version": "1.0.0",
  "tags": "flux, bfl, image-generation, prompting, t2i, i2i"
}

README badge

README badge for openmontageinc/openmontage/flux-best-practices