CONTENT CREATION

Changelog Entry with Auto-Generated Feature Screenshots

On each GitHub release, this drafts the changelog and generates a hero illustration per major feature, then publishes the illustrated entry to your changelog microsite.

CategoryContent Creation
Enginesim
Difficultyintermediate
Triggerevent
Steps6
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerGitHub release publishedGitHubGitHub
  • ActionExtract headline features from releaseGitHubGitHub
  • ActionWrite benefit-focused blurb per featureOpenAI
  • ActionGenerate illustration per featureImage generation
  • ActionAssemble and commit illustrated entryGitHubGitHub
  • OutputRedeploy microsite on VercelVercelVercel

What it does

Produces an illustrated changelog. For each GitHub release it pulls the release notes and tagged PRs, identifies the headline features, and generates a clean visual for each one so the published entry is not just a wall of bullet points. The images and copy are bundled into a single entry and deployed.

When to use it

Use it when your changelog is a marketing surface, not just an engineering log, and you want every meaningful release to ship with visuals without a designer in the loop.

How it works

  1. 1A published GitHub release fires the trigger.
  2. 2The flow extracts the headline features from the release body and linked PRs.
  3. 3An LLM step writes a short, benefit-focused blurb for each feature.
  4. 4For each headline feature it generates a matching illustration.
  5. 5The blurbs and images are assembled into one dated changelog entry and committed.
  6. 6The microsite is redeployed on Vercel with the illustrated entry live.

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 Image generationManaged Nano Banana image renders, metered per image.
  4. 4
    Connect VercelDeploys, runtime logs, analytics.
  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.