Skip to content

Deployment Changelog Generator

Example prompt: "When we push a new release tag to GitHub, compile all merged PRs since the last release, generate a changelog grouped by type, and post it to #releases on Slack."

How to automate deployment changelogs with GloriaMundo

The Problem

Release day arrives and someone has to trawl through the git log or the list of merged pull requests since the last tag, figure out which changes are features, which are bug fixes, and which are internal refactors, and then write it all up in a readable format. For teams shipping weekly or fortnightly, this changelog work takes 30-60 minutes per release and is often rushed or skipped entirely. The result is either no changelog at all, or a hastily written list that misses important changes and leaves stakeholders guessing what actually shipped.

How GloriaMundo Solves It

We build a workflow that triggers when a new release tag is pushed to GitHub. An integration step fetches all merged pull requests between the previous release tag and the new one, including PR titles, authors, labels, and descriptions. A code step processes the raw PR data to group entries by type — features, bug fixes, improvements, and internal changes — using PR labels or title prefixes as signals. An LLM step takes the grouped data and writes a polished changelog in markdown, with clear headings, concise descriptions, and contributor attributions. Finally, an integration step posts the formatted changelog to a Slack channel and optionally updates the GitHub release notes. Glass Box preview shows you the full draft changelog before it is posted anywhere.

Example Workflow Steps

  1. Trigger (webhook): Fires when a new release tag is pushed to the GitHub repository.
  2. Step 1 (integration): Fetch the previous release tag and retrieve all merged PRs between the two tags from GitHub, including titles, authors, labels, and descriptions.
  3. Step 2 (code): Group the PRs by category — features, bug fixes, improvements, and internal — based on labels and title conventions. Extract author names and PR numbers.
  4. Step 3 (LLM): Generate a formatted changelog in markdown with sections for each category, concise one-line descriptions per change, and contributor credits.
  5. Step 4 (integration): Post the changelog to #releases on Slack.
  6. Step 5 (integration): Update the GitHub release with the generated changelog as the release body.
  7. Step 6 (integration): Optionally create a Notion page with the full changelog for long-term documentation.

Integrations Used

  • GitHub — source of merged PRs, release tags, and target for release notes
  • Slack — receives the formatted changelog so the team and stakeholders see what shipped
  • Notion — optional long-term documentation of release history

Who This Is For

Engineering leads, release managers, and DevOps teams who ship regularly and want a reliable, consistent changelog without the manual overhead of writing one from scratch each time.

Time & Cost Saved

Writing a changelog manually takes 30-60 minutes per release, depending on the number of changes. For teams releasing weekly, that is 2-4 hours per month spent on changelog work alone. This workflow generates a categorised, polished changelog in seconds. The human effort is reduced to reviewing the draft in the Glass Box preview and approving it. The workflow uses integration, code, and LLM steps, costing a few credits per release.