Fal Flux Pro

This skill provides a direct pipeline to FLUX Pro v1.1 for photorealism and artistic image generation.

Install
cmdop skills install agensi-fal-flux-pro

Fal Flux Pro is an agent skill that gives an AI agent a direct pipeline to FLUX Pro v1.1, hosted on fal.ai. The skill wraps fal.ai’s async queue, handling request submission, status polling, and downloading generated images to a local directory automatically — no manual queue management required.

Once installed, an agent can generate high-resolution images from text prompts using several pre-configured aspect ratios: landscape orientations (16:9 and 4:3), square HD, and portrait. Up to four image variations can be requested in a single call, which supports iterative workflows where multiple options are needed quickly.

Safety filtering is configurable through tolerance levels ranging from 1 to 6, allowing the output to be matched to a project’s compliance requirements. The skill returns a local file path to the generated image, making it straightforward to pass the result into downstream steps in an agent workflow.

This skill is appropriate when an agent needs to produce images programmatically — for example, generating thumbnails, UI mockups, or print assets — and the workflow depends on FLUX Pro v1.1 specifically via the fal.ai platform. It is not suitable for image editing, image-to-image tasks, or scenarios where a different model or hosting provider is required. There are no listed environment variables in the facts, so check the publisher’s documentation for any fal.ai credentials or configuration needed before use.

Use cases

  • Generate photorealistic images from text prompts for YouTube thumbnails
  • Produce multiple aspect-ratio variants of a marketing image in one agent call
  • Create up to 4 image variations in a batch for rapid creative iteration
  • Generate mobile UI mockup visuals programmatically within an agent pipeline
  • Apply configurable safety tolerance levels to meet project compliance requirements
  • Download generated images to a local directory for immediate use in downstream workflow steps

When to use it

  • The agent workflow requires text-to-image generation backed specifically by FLUX Pro v1.1
  • Images need to be generated at multiple preset aspect ratios (16:9, 4:3, square, portrait)
  • The project uses fal.ai as its AI inference platform
  • Batch generation of several variations in a single call is needed
  • Async queue handling and local file management should be abstracted away from the agent logic

When not to use it

  • Image-to-image editing or inpainting is required rather than text-to-image generation
  • The project requires a model other than FLUX Pro v1.1
  • Output must be delivered to a remote storage URL rather than a local file path
  • A different inference provider than fal.ai is mandated
  • Custom or arbitrary image dimensions are needed beyond the preset aspect ratios