CONTENT CREATION

On-demand product shot cleanup via webhook with generated fallback

Accepts a product image URL over a webhook, upscales and cleans it through Replicate, and if the source is too low quality to salvage, generates a fresh studio-style image instead.

CategoryContent Creation
Enginesim
Difficultyadvanced
Triggerwebhook
Steps7
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerWebhook receives image URL and product descriptionHTTP webhook
  • ActionUpscale source and score quality with ReplicateReplicateReplicate
  • LogicBranch on quality score: cleanup vs generate
  • ActionRemove background with Replicate (cleanup path)ReplicateReplicate
  • ActionGenerate studio product image (fallback path)Image generation
  • ActionUpload final asset to Cloudflare R2CloudflareCloudflare R2
  • OutputReturn asset URL in webhook responseHTTP webhook

What it does

Exposes a single webhook endpoint that takes any product image, cleans it up to catalog spec, and — when the original is unusable — falls back to generating a fresh studio shot so the caller always gets a usable asset back.

When to use it

Use when another system (a PIM, an upload form, or an internal tool) needs to enhance a product image synchronously and get the result inline, with a graceful fallback for garbage inputs.

How it works

  1. 1A webhook receives a POST with the source image URL and product description.
  2. 2Replicate upscales the source and returns a quality/confidence score.
  3. 3A branch evaluates the score: high enough proceeds to cleanup, too low routes to generation.
  4. 4On the cleanup path, Replicate removes the background to produce a transparent PNG.
  5. 5On the fallback path, generate-image creates a clean studio-style product image from the description.
  6. 6The final asset is uploaded to Cloudflare R2 for a stable public URL.
  7. 7The webhook response returns the asset URL and which path was taken.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect HTTP webhookTrigger any URL on agent actions.
  2. 2
    Connect ReplicateImage, video, and model inference.
  3. 3
    Connect Image generationManaged Nano Banana image renders, metered per image.
  4. 4
    Connect Cloudflare R2Object storage, S3-compatible.
  5. 5
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  6. 6
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  7. 7
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.