CONTENT CREATION

On-Demand Back-Translation Check for a Submitted String Set

Accepts a batch of translated strings via webhook, back-translates each to the source language.

CategoryContent Creation
Enginesim
Difficultyintermediate
Triggerwebhook
Steps5
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerWebhook receives batch of source/target string pairsHTTP webhook
  • LogicValidate and normalize the payload
  • ActionBack-translate each target to source languageOpenAI
  • LogicScore equivalence and assign pass/flag verdict
  • OutputRespond with per-string drift resultsHTTP webhook

What it does

Exposes a webhook that takes a JSON payload of source/target string pairs (plus their locale), back-translates each target to the source language, and responds synchronously with a per-string equivalence score, a pass/flag verdict, and a short explanation of any meaning change.

When to use it

Use it as a validation endpoint for a translation vendor portal, a TMS integration, or an internal authoring tool that wants to check strings before they ever reach the repo. It is the building block other systems call rather than a scheduled or repo-driven job.

How it works

  1. 1An external system POSTs a batch of `{source, target, locale}` entries to the webhook trigger.
  2. 2The flow validates and normalizes the payload, rejecting malformed entries.
  3. 3Each target string is back-translated to its source language by the LLM.
  4. 4A scoring step computes semantic equivalence, checks placeholder and tag integrity, and assigns a pass or flag verdict with a reason.
  5. 5The webhook responds with the structured per-string results so the caller can block or accept the submission immediately.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect HTTP webhookTrigger any URL on agent actions.
  2. 2
    Connect OpenAIModels, embeddings, files.
  3. 3
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  4. 4
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  5. 5
    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.