Skip to content

moooon

Motir

Vibe your whole project. Bring an idea — Motir's three AI layers plan it, track it, and ship it, end to end. You're looking at Motir, built in Motir.

  • Vibe Project
  • Open Source
  • AI Agent
  • AI Loop
1
requests
0
upvotes
145
planned
1,361
shipped

Motir · Roadmap

MOTIR-4048To Do

Planning bug: MOTIR-3999 re-combined the publish + the re-pin on one card and omitted the 0.1.3 bump — the exact release-shape MOTIR-3965 was re-scoped out of, one release later

Opened by Zhu Yue ·

A bug/deploy card (MOTIR-3999, under epic MOTIR-3875) was authored to publish @motir/design-system@0.1.3 and re-pin motir-marketing in one card. Verified against shipped reality at claim time (2026-09-01):

  • Registry serves 0.1.2 (npm view @motir/design-system dist-tags → latest 0.1.2, published 2026-08-30T20:34:31Z).
  • motir-core origin/main packages/design-system/package.json reads 0.1.2no 0.1.3 bump exists anywhere (no card, no open PR, no merged commit; git log origin/main -- packages/design-system/package.json last touches the 0.1.2 bump b32ca9c0e).
  • .github/workflows/release-design-system.yml fires on a design-system-v<x.y.z> tag and fails fast unless the tag equals package.json's version; packages/design-system/README.md § Releasing states bump → merge → tag.
  • The card's own Boundary says "No source change to either".

So criterion 1 ("published at the next patch version above the registry") reads on a human merge (the 0.1.3 bump) that no run in manual merge mode can produce, and there is no bump card for the edge to carry. Criteria 3–6 are discharged in a second repository (motir-marketingpackage.json, pnpm-lock.yaml, tests/aaMatrix.test.ts, the built /design page) whose lockfile is unobtainable until the publish happens — a strictly-ordered two-repo straddle, the exact carve-out exclusion MOTIR-3965's halt established.

This is the same shape MOTIR-3965 was re-scoped out of on 2026-08-30 (plan cmtfw2fp3006khwph9ycw7qtp → bump MOTIR-3995 / cut MOTIR-3965 / re-pin MOTIR-3996), and the global lesson recorded on MOTIR-3992 ("A card that publishes an artifact and then consumes it spans a merge no run can perform…") — reinforced with occurrenceRef MOTIR-3999 on 2026-09-01 (occurrences: 2). The card was authored AFTER the re-scope and the lesson, and still recombined the publish + re-pin and omitted the bump. validate_work_item returned valid: true with no repo-straddle advisory (the straddle detector reads criteria for a PATH in another repo; these name the repo by bare name).

Action taken

Corrected shape submitted as plan cmth8rbd60016hvn8712x9fo1 (2026-09-01), one card per merge boundary:

  • ADD task (motir-core, chore, 1pt/15m) — bump packages/design-system/package.json to 0.1.3. One file, one PR.
  • MODIFY MOTIR-3999 — re-scoped to the CUT alone (deploy, 1pt/25m): push design-system-v0.1.3 on the merged bump commit, watch the lane, pull the tarball anonymously, verify the @scope form in the tarball, quote the shasum. blocked_by the bump.
  • ADD task (motir-marketing, code, 2pt/30m) — re-pin to 0.1.3, regenerate the lockfile, verify the STYLE gallery on the built site under glassmorphism. blocked_by MOTIR-3999.

Nothing archived; MOTIR-3999 survives as the middle step, keeps its links and provenance. motir run MOTIR-3999 halted at the replan and did not build.

**⚠️ Also recorded (2026-09-01): a TENANT lesson "A card that publishes an artifact and then re-pins its consumer spans a merge no run can cross…" was written via add_lesson (id cmth8phpo000dhvskhh6cs6ol) BEFORE the correct recurrence instrument was applied — it duplicates the existing GLOBAL lesson cgycb03naytbpb1fwxieicuvk (sourceRef MOTIR-3992) and should be retired via the Lesson Library; the global row carries the reinforcement.

Discussion

No comments yet.

Adding to this discussion signs you in on app.motir.co and brings you back to this request.

Add a comment