PlaysbyMoat8
Back to the library
The workflow guideProduct & Engineering

Draft product updates from code changes

Turn merged code changes into concise product updates that explain the user-facing features and fixes.

5 implementation stepsIntermediate

How this works

Turn merged code changes into concise product updates that explain the user-facing features and fixes. Use this to draft understandable product updates from a defined set of code changes. The draft should describe the user-visible result while distinguishing merged work from functionality actually released to the audience.

Gather merged changes for a release window, inspect relevant descriptions and diffs, then verify release and feature-flag state. Group related changes into concise entries with internal evidence links. Engineering and product owners confirm availability and wording before publication.

Before you start

Release scope

Identify the version, date range and target environment.

Change evidence

Obtain merged pull requests, relevant commits and issue context.

Availability evidence

Use deployment records and feature-flag status for the intended audience.

Prepare the inputs

change_set
Required

PR ID, merge commit, title, description and diff

Exclude unmerged or reverted changes unless explaining a rollback.

release_record
Required

version, deployed commit and environment

A merge timestamp alone does not establish shipping.

audience
Required

customer segment and communication format

Avoid internal implementation detail unless it helps users.

editorial_rules
Required

entry categories and disclosure constraints

Keep security-sensitive or unreleased details out of public drafts.

Implementation, step by step

  1. Establish the release boundary

    • Map the deployed version to included commits and changes.
    • Identify reversions and changes behind inactive flags.

    Checkpoint: A candidate change list has availability status.

  2. Inspect user impact

    • Read descriptions and relevant diffs to identify changed behavior.
    • Separate internal maintenance from visible features and fixes.

    Checkpoint: Each candidate has an evidence-backed impact statement.

  3. Group related changes

    • Combine several PRs that implement one user-facing improvement.
    • Avoid counting refactors as new features or omitting a consequential compatibility change.

    Checkpoint: The outline contains distinct product outcomes.

  4. Draft entries

    • Explain what users can now do or what failure was fixed.
    • Use plain language and keep internal source links for reviewer verification.

    Checkpoint: The draft matches the intended release audience.

  5. Verify and hand off

    • Ask owners to confirm availability, claims and any limitations.
    • Save the approved draft in the requested format; publication is a separate authorized action.

    Checkpoint: Every entry has a verified release basis.

Decisions along the way

Merged but not deployed

Exclude from shipped notes or label as upcoming only if requested.

Feature is enabled for a subset

State the actual availability or use a targeted draft.

Change has no clear user effect

Keep it in internal notes rather than invent a benefit.

What you will produce

release-notes.md

Markdown

Customer-oriented release notes awaiting review.

Include: version · features · fixes · known_limits · status

changelog-evidence.csv

CSV

A traceability table for each published claim.

Include: entry_id · pr_ids · commit · availability · owner

Check the result

  • Confirm every entry is included in the deployed version.
  • Check feature flags and audience availability.
  • Compare claimed behavior with code and relevant validation evidence.
  • Remove internal secrets, customer identifiers and unsupported performance claims.

When something goes wrong

Deployment record unavailable

Keep availability unverified and hold shipped wording.

PR description contradicts the diff

Inspect the implementation and ask the owner to clarify.

Late rollback occurs

Update affected entries and retain the prior draft for history.

A worked example

SYNTHETIC EXAMPLE · NOT A REPORTED BUSINESS RESULT

Request

Synthetic: draft notes for release R6.

Inputs

PR11 adds saved filters and is deployed with its flag enabled. PR12 adds export but its flag is disabled. PR13 refactors logging with no visible change.

Expected result

Synthetic result: include saved filters, omit export from shipped features and keep logging in internal maintenance notes. The evidence table links the entry to PR11 and R6.

Run a small pilot

Start small

Prepare notes for one recent release with known deployed behavior.

Record the baseline

Record editing time and availability corrections.

Agree success criteria

Agree factual accuracy and user clarity criteria before recurring generation.

Stop or revise

Stop publication preparation if deployed commits or feature availability cannot be established.

Keep the scope clear

  • Generated notes do not prove deployment or feature availability.
  • Do not publish or disclose unreleased implementation details without the authorized scope.

Check before using

Check that features are actually available to users. A merged pull request does not establish that a release has shipped.

Business activities

  • Create and publish content
  • Build and change software

Editorially adapted workflow; not independently tested.

FROM A RECIPE TO YOUR WORKFLOW

Bring the company context together.

This workflow needs reliable inputs before it needs a better prompt. Moat8 Vault brings working information into a readable archive for the AI tools you use.

For this recipe: Merged pull requests for a period · Product communication style

Start with one recurring task, agree which sources to use, and review the first result with your team.

Discuss this workflow

What does your team do?

Choose the recurring work you would like to improve.

A field guide, made for work.

Plays is a curated library of anonymized business workflows. Each recipe explains the inputs, steps, tools, and checks you can adapt to your team. These are editorial guides, not independently tested implementations.

When you share a company website, the guide may read a limited number of public pages on that website. Messages and retrieved page text are sent to our AI provider, NVIDIA, to recommend relevant workflows. Share a general description rather than confidential information.

Saved workflows and selected activities stay in this browser. Your chat history and selected company website are kept in this tab and cleared when you reload or start a new chat. The server does not store conversations. Live source monitoring is not enabled.

If you allow analytics, PostHog receives page views and actions such as workflow saves, skill downloads and link clicks. Random browser and session identifiers connect these actions without identifying you by name. Your choice and browser identifier expire after 90 days; sessions restart after 30 minutes of inactivity or 24 hours. We use local storage, not analytics cookies. Messages, company URLs, search text, full referrers and IP addresses are not sent. There are no person profiles or session recordings. You can decline or withdraw in Analytics settings; Do Not Track and Global Privacy Control keep analytics off.

Read the Plays Privacy Notice ↗