Week one of the gitpulse experiment is in the books. The homepage now surfaces release editions alongside pull request features, and a dedicated /releases/ section gives every tagged release its own detail page with changelogs and shareable links. Story URLs mirror GitHub's structure—pull requests land at /pull/<n>/ and commits at /commit/<sha>/—so every link is instantly meaningful.
The project also shipped its first test suite: 53 vitest tests across the action and site, covering the workflows that turn raw git history into editorial copy. A complete README install guide means new users can wire gitpulse to their own repositories without asking for help. First-run state now guides newcomers to the setup docs instead of dropping them into an empty feed.
Under the hood, releases are tracked as first-class entities with validation and storage. The analyzer pipeline processes them end-to-end, matching each release to its constituent PRs and generating AI-written copy. Hash-based caching prevents redundant LLM calls on unchanged data, keeping costs predictable as the repository grows.
The homepage now displays release editions above PR features, and a new /releases/ section lists every release with hero cards, compact rows, and full detail pages with changelogs.
Gitpulse now ships its first test suite — 53 vitest tests across the action and site workspaces — alongside a complete README install guide with provider examples and a friendlier first-run state that guides new users to the setup docs.
The analyzer pipeline now processes GitHub releases end-to-end, matching each release to its constituent PRs and generating AI-written editorial copy. Hash-based caching prevents redundant LLM calls on unchanged data.