#450 feat(dashboard): add a What's New changelog + roadmap modal

[Anonymous]

Originally created by: Akarsh-Hegde

Summary

• Adds a "What's New" modal to the dashboard: release history (highlights + fixes) and a roadmap, sourced from a small hand-curated tray/src-tauri/resources/whats-new.json — deliberately separate from the auto-generated CHANGELOG.md, which is commit-level and too internal to show end users.
• Auto-opens once per app version after an update lands, via the same poll-tick pattern as the existing "Plan your day" auto-open. Defers itself whenever the dashboard window is already open, so it never collides with the sibling Plan auto-open on the same tick and never yanks the user off whatever they're doing.
• Always reachable via a new "What's New" item in the toolbar nav pill.
• Adds a "What's New per release" recipe to CLAUDE.md's Common Tasks so the content doesn't drift out of sync with real releases.

Note for reviewers

The seeded roadmap content ("Idle & sleep detection", "Lighter local AI footprint") is a starting placeholder pulled from documented known-limitations — please confirm or replace it before this reaches users, since it reads as a product-facing promise.

Test plan

• [x] cargo fmt --check / cargo clippy -- -D warnings / cargo test (tray crate) — all green, including new whats_new unit tests (JSON-shape guard + seen/unseen marker logic)
• [x] npm run build (ui) — TypeScript + static export clean
• [x] Pre-push hooks (fmt, clippy, ui build/tests, cargo test, security audit) — all passed
• [x] npm run tauri dev manual pass — auto-open firing, modal render, tab switching, nav-pill open, close-marks-seen (not exercised by CI or this session — WKWebView isn't automatable in this environment)

🤖 Generated with Claude Code

https://claude.ai/code/session_01MrJAJ3nnr5C1iyaKGV3BiY

[Anonymous]

Originally posted by: coderabbitai[bot]

[!IMPORTANT]

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: f38370b4-2357-46fb-a259-e62e732482bf

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:
- [ ] 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

- [ ] Create PR with unit tests - [ ] Commit unit tests in branch `feat/whats-new-changelog`

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

- [X](https://twitter.com/intent/tweet?text=I%20just%20used%20%40coderabbitai%20for%20my%20code%20review%2C%20and%20it%27s%20fantastic%21%20It%27s%20free%20for%20OSS%20and%20offers%20a%20free%20trial%20for%20the%20proprietary%20code.%20Check%20it%20out%3A&url=https%3A//coderabbit.ai) - [Mastodon](https://mastodon.social/share?text=I%20just%20used%20%40coderabbitai%20for%20my%20code%20review%2C%20and%20it%27s%20fantastic%21%20It%27s%20free%20for%20OSS%20and%20offers%20a%20free%20trial%20for%20the%20proprietary%20code.%20Check%20it%20out%3A%20https%3A%2F%2Fcoderabbit.ai) - [Reddit](https://www.reddit.com/submit?title=Great%20tool%20for%20code%20review%20-%20CodeRabbit&text=I%20just%20used%20CodeRabbit%20for%20my%20code%20review%2C%20and%20it%27s%20fantastic%21%20It%27s%20free%20for%20OSS%20and%20offers%20a%20free%20trial%20for%20proprietary%20code.%20Check%20it%20out%3A%20https%3A//coderabbit.ai) - [LinkedIn](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fcoderabbit.ai&mini=true&title=Great%20tool%20for%20code%20review%20-%20CodeRabbit&summary=I%20just%20used%20CodeRabbit%20for%20my%20code%20review%2C%20and%20it%27s%20fantastic%21%20It%27s%20free%20for%20OSS%20and%20offers%20a%20free%20trial%20for%20proprietary%20code)

_{Comment @coderabbitai help to get the list of available commands.}

[Anonymous]

Originally posted by: adityaharishch

Review: correctness-focused pass

Overall implementation is solid — no crashes or races found in the primary paths.

1. "Seen" tracking (whats_new.rs:71-90, whats_new_auto_open.rs:71) — correct. Persisted to ~/.meridian/last_seen_version (survives restart), exact-string-compared against app.package_info().version. Since app versions are always exact per release, this can't loop on a downgrade/rebuild, and an absent/mismatched file both correctly read as "unseen."

2. Auto-open deferral vs. sibling "Plan your day" (poll/mod.rs:141,146) — correct, not racy. Both run sequentially (.await) on the same tick; whats_new_auto_open runs strictly after plan_auto_open and checks get_webview_window("dashboard").is_some(), so if Plan opens the dashboard first, What's New correctly sees it and defers.

3. JSON parsing (whats_new.rs:57-61) — build-time malformation is caught by the whats_new_json_parses unit test, but at runtime parse_whats_new() uses .expect(...). If a malformed whats-new.json ever shipped (hooks bypassed, hand-edit, etc.), the first get_whats_new() call (on modal open) would panic instead of degrading. Worth a serde_json::from_str(..).unwrap_or_else(|_| WhatsNewData { releases: vec![], roadmap: vec![] }) fallback for defense-in-depth even though the test gate makes this low-risk in practice.

4. Command/window wiring — fine, get_whats_new/mark_whats_new_seen_cmd registered in lib.rs's invoke_handler!, reuses the existing "dashboard" window capability — no gap.

5. Seen-marking timing — minor inconsistency, not a bug: the auto-open path (whats_new_auto_open.rs:71) marks the version seen before the window renders (documented as an intentional crash-safety tradeoff), which is subtly at odds with the doc comment at whats_new.rs:92-94 claiming seen "always reflects the version actually viewed." A crash between the marker write and the window rendering means that version's notes are silently never shown again (toolbar pill is the only remaining path to them) — worth tightening the comment to acknowledge this, even if the tradeoff itself is fine.

6. Types — Rust/TS types match, except RoadmapItem.status is an unvalidated String in Rust vs. a TS literal union. A typo'd status in whats-new.json (e.g. "Planned" vs "planned") parses fine on the Rust side but silently renders undefined label/color via the ROADMAP_STATUS_LABEL/_COLOR lookups (WhatsNewModal.tsx:139,147) instead of erroring loudly.

No blocking issues — nice work on the deferral logic against the sibling auto-open.

[Anonymous]

Originally posted by: Akarsh-Hegde

Addressed the review findings from @adityaharishch's correctness pass:

• #3 (panic risk) — parse_whats_new() no longer .expect()s on a malformed embedded JSON. It now logs tracing::error! and serves an empty {releases: [], roadmap: []} payload, so a bypassed hook or hand-edit degrades the modal gracefully instead of panicking on first open.
• #6 (unvalidated status) — RoadmapItem.status is now a #[serde(rename_all = "kebab-case")] enum (InProgress | Planned | Considering) instead of a bare String. A typo'd value in whats-new.json now fails the JSON parse loudly (and falls into the same empty-payload fallback above) instead of silently rendering undefined label/color on the frontend lookups. No frontend change needed — the wire format is unchanged.
• #5 (doc accuracy) — tightened mark_whats_new_seen_cmd's doc comment to acknowledge that the auto-open path already writes the "seen" marker before the window renders, rather than implying this call is always what marks a version seen.

Items [#1], [#2], and [#4] were reviewed and confirmed correct as-is, no change needed.

Also pushed the visual polish pass on WhatsNewModal.tsx (accent bars, a "Latest" badge, colored bullet dots, staggered pop-in) that was finished but not yet committed from earlier manual testing — data/logic unchanged.

Verified: cargo fmt --check, cargo clippy -- -D warnings, cargo test (69/69 passing, including the new/existing whats_new tests), and npm run build in ui/ all clean. Full pre-push suite (fmt, clippy, ui build + tests, cargo test, security audit) passed on push.

[Anonymous]

Ticket changed by: Akarsh-Hegde

• status: open --> closed