fix: preserve approved Featured order across public catalogs

[Patrick-Erichsen]

What Problem This Solves

Public Featured feeds currently order by badge timestamps, so they can display a different sequence from the approved staff selection or truncate it before ordering.

User Impact

Plugin and skill Featured feeds preserve the saved publication order before applying limits. Membership and eligibility still come from the canonical badge and public-artifact owners; unavailable entries never become public cards. Existing manually featured entries without a saved publication retain deterministic legacy order.

Why This Change Was Made

The shared ordering helper is used by plugin pages/overview and the native skill feeds, including paginated skill catalog access. It does not let later editorial edits reorder an already published selection. Long publisher labels also shrink correctly on mobile so the artifact title remains visible.

Evidence

A same-input native regression fails on the previous badge order and passes with the exact 16 approved identities through plugin page, skill V4, highlighted list, listWithLatest, and limit-one skill catalog pagination. Real local browser proof shows the publication order in both catalogs across four viewport widths. The mobile title improves from 0px to 72.7px on identical input.

The full combined stack passed 7,070 tests (3 skipped), static/audit, types/build, package and HTTP/CLI checks, explicit Convex/package type checks, and 19 browser smoke cases. Shared safe proof is linked below after publication.

Production +89/-33 (net +56); tests +224/-92. The added owner helper makes saved order canonical across public readers; no new ranking or publication path is introduced.

Real browser and native publication proof.

[clawsweeper]

🦞👀
ClawSweeper picked this up.

Pull request received. I will update this pull request when review starts.

ClawSweeper review complete

ClawSweeper finished reviewing this revision. The review result is being finalized.

View the workflow run.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated
clawhub	Ready	Preview	Sep 16, 2026 8:45pm UTC

[clawsweeper]

Codex review: needs maintainer review before merge. Reviewed September 16, 2026, 4:47 PM ET / 20:47 UTC.

ClawSweeper review

What this changes

Preserve approved Featured ordering across public skill and plugin catalogs, retain that order through homepage filtering, and keep titles visible beside long publisher names.

Merge readiness

✅ Ready for maintainer review

Keep open. The fix remains necessary on current main, and the reviewed changes have no blocking findings. This collaborator-authored PR complements its open parent rather than duplicating a merged fix.

Priority: P2
Reviewed head: c40837e7cf33502d7fa810cbae11eb749c8004e3

Review scores

Measure	Result	What it means
Overall readiness	🐚 platinum hermit (4/6)	A focused implementation with useful real-setup evidence and regression coverage, with no blocking defect found.
Proof confidence	🐚 platinum hermit (4/6) ✨ media proof bonus	Sufficient (linked_artifact): The linked real React/disposable-Convex report exercises public Featured readers and records both catalogs in approved order; inspected mobile before/after screenshots demonstrate the CSS repair. Evidence uses a local combined-stack revision, with the independent source-identity limitation recorded above.
Patch quality	🐚 platinum hermit (4/6)	No actionable review findings were identified.

Verification

Check	Result	Evidence
Real behavior	Verified	Sufficient (linked_artifact): The linked real React/disposable-Convex report exercises public Featured readers and records both catalogs in approved order; inspected mobile before/after screenshots demonstrate the CSS repair. Evidence uses a local combined-stack revision, with the independent source-identity limitation recorded above.
Evidence reviewed	7 items	Introduced scope verified: The pinned base-to-head delta contains 12 files: production +89/-33 and tests +224/-92. The publication schema and staff publication workflow belong to the parent branch, not this introduced delta. Current main still needs the fix: Current main fetches Featured packages in badge order and slices that sequence; the homepage also describes and uses badge-timestamp ordering. The saved-selection helper is absent from current main and the supplied latest-release commit v0.23.3. Ordering preserves membership boundaries: The shared helper sorts existing candidates using published identities. Package readers retain visibility and category checks; the new skill catalog branch retains skillCatalogMatchesFilters and public-version hydration. No authorization, publication mutation, dependency, or workflow change is introduced.
Findings	None	None.
Security	None	None.

How this fits together

ClawHub’s Featured catalogs combine staff-selected membership with public artifact eligibility. These readers turn eligible badges and saved publication order into catalog responses and homepage cards.

flowchart TD
  A[Featured badges] --> C[Public eligibility checks]
  B[Saved publication order] --> D[Order eligible entries]
  C --> D
  D --> E[Catalog limits and pagination]
  E --> F[Homepage category filtering]
  F --> G[Featured cards]

Loading

Before merge

None.

Agent review details

Security

None.

Review metrics

Metric	Value	Why it matters
Change size	+313/-125 across 12 files	The patch is bounded to public catalog ordering, homepage presentation, and associated tests.
Production versus tests	Production +89/-33; tests +224/-92	Production growth is justified by sharing publication ordering across existing public readers.

Technical review

Best possible solution:

Keep the saved publication snapshot as the shared ordering source while existing badges, visibility checks, and legacy fallback continue to govern eligible results.

Do we have a high-confidence way to reproduce the issue?

Yes: differing badge timestamps and approved identity order expose the mismatch in the existing readers, and the supplied regression fixture exercises that case. This review did not execute a current-main reproduction.

Is this the best way to solve the issue?

Yes, this is the best layer for the fix: shared server-side ordering covers API consumers as well as the homepage. Frontend-only sorting would leave other readers inconsistent, while rewriting badge timestamps would alter unrelated provenance.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning medium; reviewed against 7725f32bf07d.

Labels

Label changes:

• add P2: This repairs public catalog ordering and mobile title visibility with a limited discovery-surface impact.
• add proof: sufficient: Contributor real behavior proof is sufficient. The linked real React/disposable-Convex report exercises public Featured readers and records both catalogs in approved order; inspected mobile before/after screenshots demonstrate the CSS repair. Evidence uses a local combined-stack revision, with the independent source-identity limitation recorded above.
• add rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🐚 platinum hermit and patch quality is 🐚 platinum hermit.
• add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (linked_artifact): The linked real React/disposable-Convex report exercises public Featured readers and records both catalogs in approved order; inspected mobile before/after screenshots demonstrate the CSS repair. Evidence uses a local combined-stack revision, with the independent source-identity limitation recorded above.

Label justifications:

• P2: This repairs public catalog ordering and mobile title visibility with a limited discovery-surface impact.
• rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🐚 platinum hermit and patch quality is 🐚 platinum hermit.
• status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (linked_artifact): The linked real React/disposable-Convex report exercises public Featured readers and records both catalogs in approved order; inspected mobile before/after screenshots demonstrate the CSS repair. Evidence uses a local combined-stack revision, with the independent source-identity limitation recorded above.
• proof: sufficient: Contributor real behavior proof is sufficient. The linked real React/disposable-Convex report exercises public Featured readers and records both catalogs in approved order; inspected mobile before/after screenshots demonstrate the CSS repair. Evidence uses a local combined-stack revision, with the independent source-identity limitation recorded above.

Evidence

What I checked:

• Introduced scope verified: The pinned base-to-head delta contains 12 files: production +89/-33 and tests +224/-92. The publication schema and staff publication workflow belong to the parent branch, not this introduced delta. (c40837e7cf33)
• Current main still needs the fix: Current main fetches Featured packages in badge order and slices that sequence; the homepage also describes and uses badge-timestamp ordering. The saved-selection helper is absent from current main and the supplied latest-release commit v0.23.3. (convex/packages.ts, 7725f32bf07d)
• Ordering preserves membership boundaries: The shared helper sorts existing candidates using published identities. Package readers retain visibility and category checks; the new skill catalog branch retains skillCatalogMatchesFilters and public-version hydration. No authorization, publication mutation, dependency, or workflow change is introduced. (convex/lib/featuredSelections.ts:68, c40837e7cf33)
• Focused regression coverage: The added convex-test fixture exercises five public readers, manual entries outside the snapshot, an unbadged reservation, and limit-one skill catalog pagination. Homepage tests cover category filtering without reordering; legacy fixtures explicitly model an absent snapshot. Tests were inspected, not executed during this read-only review. (convex/featuredPublicOrder.runtime.test.ts:126, c40837e7cf33)
• Real setup evidence inspected: The linked report and summary describe real React, Playwright, and disposable Convex execution: both public catalogs returned 16 approved identities in order, pending reservations produced no cards, and four viewport widths were checked. Directly inspected mobile before/after screenshots show the title becoming visible; recorded width changes from 0 to 72.734375 pixels. Summary blob identity: e570bbec0a197d1317a933d3dddfc68d4d373636. The artifact records local candidate ce75933cdd; GitHub could not resolve that local revision, so byte-for-byte equivalence with this PR head was not independently established. (clawhub-ui-proof/pr-3765/featured-sixteen/summary.json, a8a7faae7149)
• Open parent supplies publication ownership: feat: review and publish 16-item Featured selections from monthly installs #3765 remains open and unmerged, with head equal to this PR’s pinned base. It supplies saved 16-item publication snapshots; this PR supplies their public ordering. The canonical search found no verified merged replacement for this work. (e8fe2f5a82e0)

Likely related people:

• Patrick-Erichsen: Suggested for follow-up; no historical authorship or introduction is verified. (role: unverified routing candidate; confidence: low)
• vyctorbrzezowski: Suggested for follow-up; no historical authorship or introduction is verified. (role: unverified routing candidate; confidence: low)

Rating scale

Score	Internal tier	Crab rank	Meaning
6/6	S	🦀 challenger crab	Exceptional readiness
5/6	A	🦞 diamond lobster	Very strong readiness
4/6	B	🐚 platinum hermit	Good normal PR; ordinary maintainer review
3/6	C	🦐 gold shrimp	Useful, but confidence is limited
2/6	D	🦪 silver shellfish	Proof or implementation needs work
1/6	F	🧂 unranked krab	Not merge-ready
N/A	NA	🌊 off-meta tidepool	Rating does not apply

Overall follows the weaker of proof and patch quality.
Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

Workflow

• ClawSweeper keeps one durable marker-backed review comment per issue or PR.
• Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
• A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
• PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
• Maintainers can also comment @clawsweeper review to request a fresh review only.
• Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
• Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
• Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.