CONTENT CREATION
Extract and translate on-screen text before localizing
Runs OCR on a product screenshot via a Hugging Face model to read the embedded UI text, translates each string, and rebuilds the screenshot with localized copy in place.
How it runs
The automated pipeline, trigger to output.
- TriggerOperator submits screenshot + target locale
- ActionOCR the screenshot text via Hugging FaceHugging Face
- LogicRoute low-confidence reads to manual review
- ActionRegenerate screenshot with translated copyImage generation
- OutputUpload localized image to R2 and return reportCloudflare R2
What it does
Handles screenshots where the marketing copy is part of the image itself. It uses a Hugging Face OCR model to read the visible UI text, translates each detected string, and regenerates the screenshot with the localized text composited back into the same layout.
When to use it
When you only have a flat PNG export with no editable source layer and you still need accurate localized versions — for example reusing partner or legacy screenshots whose design files are gone.
How it works
- 1A manual trigger supplies the screenshot URL and the target locale.
- 2A Hugging Face OCR inference call returns the detected text regions and their bounding boxes.
- 3A logic step decides whether enough text was found to localize; low-confidence reads are routed to a manual-review flag instead.
- 4Each detected string is translated into the target locale.
- 5The screenshot is regenerated with the translated strings placed back into their original regions.
- 6The localized image is uploaded to R2 and the URL plus the extracted-text report are returned.
Set it up
What you configure once, before turning it on.
- 1Connect Hugging FaceModels, datasets, spaces — the open-source hub.
- 2Connect Image generationManaged Nano Banana image renders, metered per image.
- 3Connect Cloudflare R2Object storage, S3-compatible.
- 4Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
- 5Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
- 6Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.
More Content Creation workflows
Build a Dropbox B-roll library from exported blog drafts
Triggers when a new blog draft file lands in a Dropbox folder, generates a Replicate B-roll clip for each paragraph.
Generate B-roll clips when a blog post is marked Ready in Notion
Watches a Notion blog database for posts flipped to 'Ready', splits the body into paragraphs, and generates a short B-roll video clip for each paragraph with Replicate.
On-demand: turn a blog URL into narrated B-roll clips
Accepts a blog URL via webhook, scrapes the article, generates a B-roll clip per paragraph with Replicate and a voiceover with ElevenLabs.
On-Demand Screenshot Freshness Report from a Slack Command
A writer requests a freshness check in Slack for a specific docs page; the workflow inspects every screenshot on that page, scores each by age and version gap.
Sync a Figma master frame into localized hero renders
Pulls the current master hero frame and its text layers from Figma, generates a localized image variant for each locale defined in Airtable.
Research-driven localized hero banners tuned per market
An agent researches cultural and seasonal cues for each target market, adapts the master hero concept to fit local context, generates a tailored image per locale.
Run it inside a business
This workflow drops into a full company template. Import the org, and this is one of the playbooks its agents run.

Run this workflow in your colony.
14-day trial. No DevOps. No Sales call. Provisioned in under a minute.
