Skip to content
Folder Tag Sync Folder Tag Sync Folder Tag Sync

Submission PR stalled

What happened

Section titled “What happened”

PR #8796 (Add plugin: Folder Tag Sync) to obsidianmd/obsidian-releases has been open since 2025-12-01. It’s gone through:

  • 7 bot scans, progressively narrowing the required fixes
  • Final sentence-case violations fixed in commit ffe01d5 (2025-12-07)
  • Close/reopen performed on 2025-12-08 and again on 2026-01-14 to force rescan
  • User comment “Not stale, just waiting” on 2026-01-14
  • User comment “Still waiting” on 2026-03-13

As of today (2026-04-13): Bot’s last scan is still from 2025-12-06 on commit e1b1d2f — which is before the fix commit. The bot has never rescanned the fixed code despite multiple close/reopens.

Stale-bot warnings on 2026-01-12, 2026-02-26, 2026-04-13. Next auto-close ~2026-04-28.

Why it matters

Section titled “Why it matters”

The plugin is code-complete and published to GitHub with the 0.1.7 release. Users can install via BRAT. But without community plugin listing, discoverability is basically zero.

All the remaining bot-flagged issues are fixed and verified locally with eslint-plugin-obsidianmd. We’re not blocked on code — we’re blocked on:

  1. The ObsidianReviewBot actually rescanning the latest code
  2. A human reviewer from the Obsidian team approving

Neither is something we can force.

What to try next

Section titled “What to try next”

Keep the PR alive

Section titled “Keep the PR alive”

Bump it before each 30-day staleness warning (comment or close/reopen). Loses nothing.

Fresh PR

Section titled “Fresh PR”

Close this one and submit a new PR. Might skip the queue, or might land in the same queue further back. Risk.

Tag a reviewer

Section titled “Tag a reviewer”

Reluctantly. The team actively discourages this. Last resort.

Nothing

Section titled “Nothing”

Let auto-close happen, resubmit when we have more features or screenshots. Least friction, longest timeline.

What we learned

Section titled “What we learned”
  • ObsidianReviewBot is unreliable — claimed “6-hour rescan” never happened in our case
  • Close/reopen is documented as a rescan trigger but doesn’t always work
  • The plugin review queue is measured in months, not weeks
  • Stale-bot auto-close at 30 days creates a reminder cadence — use it
  • Local eslint-plugin-obsidianmd catches everything the bot catches. Run it before every push.
Section titled “Related work”
  • Plugin code is done. Phase 2 (conditional form fields, rule packs) is on hold pending acceptance.
  • Docs site (this very KB) built today as a way to make the plugin look more polished when a reviewer eventually looks.