feat(release): add make package-desktop for claude.ai desktop plugin upload

[yyiilluu]

What

Adds a one-command, leak-proof way to update the Claude desktop app to a new claude-smart release.

• scripts/build-desktop-plugin.sh — builds an uploadable plugin .zip from the npm tarball's vetted file set, so it matches exactly what npm ships (no .venv / node_modules / build-cache leakage). Prints the artifact path on stdout, instructions on stderr. Flags: --skip-build (reuse newest existing tarball), --output PATH. Self-checks that the manifest is at the zip root and no runtime caches slipped in.
• make package-desktop — depends on package (fresh tarball), then runs the script → dist/claude-smart-desktop-<version>.zip. Added to .PHONY and make help.
• DEVELOPER.md — new "Desktop app (claude.ai) — manual plugin upload" section under Release flow, plus a pre-release checklist bullet.

Why

The Claude desktop app (Cowork / "local agent mode") installs plugins from a claude.ai marketplace, and a marketplace can only sync from git. This repo is deliberately non-installable from git — npm-only distribution, with the vendored Reflexio runtime (plugin/vendor/reflexio) and the marketplace manifest gitignored and generated at pack time. So the desktop app cannot track npm releases automatically; updating it is a manual Upload plugin of a built bundle. Producing that bundle by hand is error-prone (zipping plugin/ in place sweeps in .venv + node_modules + .next, ballooning to hundreds of MB). This makes it a single, correct command.

Testing

• make package-desktop wiring verified; the script produces a ~2.6 MB zip with .claude-plugin/plugin.json at the archive root, the Reflexio runtime vendored in, and no .venv / node_modules.
• --help, --skip-build, and --output exercised; make help lists the new target.
• Output goes to the already-gitignored dist/; no new ignore rules needed.

[coderabbitai]

Warning

Review limit reached

You’ve reached a temporary PR review limit under our Fair Usage Limits Policy.

Your recent review volume is higher than typical usage, so adaptive limits are currently applied.

Next review available in: 33 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 5d464a72-15ce-417b-a743-edf06b68333a

📥 Commits

Reviewing files that changed from the base of the PR and between f73a8ef and 086726d.

📒 Files selected for processing (3)

• DEVELOPER.md
• Makefile
• scripts/build-desktop-plugin.sh

---

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