CONTENT CREATION

Fan out merged changelog entries to per-locale draft notes

When a changelog file is updated on the main branch, translate each new entry into your target locales and post the drafts to a review channel for human approval before publishing.

CategoryContent Creation
Enginesim
Difficultyintermediate
Triggerevent
Steps5
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerPush to default branch touches CHANGELOGGitHubGitHub
  • ActionDiff changelog and extract newly added entriesGitHubGitHub
  • LogicStop if no customer-facing entries were added
  • ActionRewrite and translate each entry per localeOpenAI
  • OutputPost localized drafts to Slack review channelSlack

What it does

Watches your repository's CHANGELOG for merges to the default branch, extracts the newly added entries, and produces a customer-facing release note in each of your configured locales. Every translation lands in a review channel as a draft — nothing reaches customers until a human approves it.

When to use it

Use this when engineering merges raw, terse changelog lines ("fix: null pointer in invoice export") and you need polished, localized customer notes without manually rewriting and translating each one. Ideal for teams shipping to multiple regions who keep a single source-of-truth changelog.

How it works

  1. 1A push to the default branch that touches the changelog file triggers the flow.
  2. 2The flow diffs the changelog and isolates only the lines added in this push.
  3. 3If no customer-relevant entries were added (e.g. only internal/chore lines), the run stops.
  4. 4An LLM rewrites each entry into friendly customer copy, then translates it into every target locale.
  5. 5Each localized draft is posted to a Slack review channel with approve/reject context and a link back to the merge.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect GitHubRepos, issues, pull requests, actions.
  2. 2
    Connect OpenAIModels, embeddings, files.
  3. 3
    Connect SlackChannels, DMs, threads, mentions.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    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.