Migrate Tags

SkillFiles & storage

Migrate body-level #hashtags to frontmatter `tags:` lists so Bases can filter on them. Idempotent — leaves files that already have frontmatter tags alone.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Migrate Tags skill

What this skill tells your AI

The instructions your AI receives, as published by benoror/obsidianos_work in .agents/skills/migrate-tags/SKILL.md and read by ahel’s review.

Usage

  • /migrate-tags — Scan vault, present a migration plan, prompt to confirm.
  • /migrate-tags <path> — Migrate a single file.
  • /migrate-tags <folder> — Migrate all .md files under a folder.
  • /migrate-tags --dry-run — Show what would change without writing.

Why this skill exists

Obsidian supports two tag styles:

  • Body tags: #frontend written inline in note text.
  • Frontmatter tags: a YAML tags: list at the top of the file.

Bases can filter on the frontmatter style (file.hasTag("frontend")) but not the body style reliably. Migrating one to the other makes tags queryable in Bases views — e.g. a "People with #mobile tag" view becomes trivial once everyone's @Name.md has tags: [mobile].

This skill is a one-shot migration utility plus an ongoing cleanup tool for new notes.

Workflow

Step 1: Discover candidates

Use obsidian tags format=json to enumerate all tags + their usage counts and source files. Identify files where:

  • A body #tag exists, AND
  • The same tag is NOT already present in frontmatter tags:.

These are the migration candidates.

Step 2: Present the plan

Show a numbered table grouped by file:

| # | File | Body tags to migrate | Existing frontmatter tags |
|---|------|----------------------|---------------------------|
| 1 | Teams/People/@Jane Doe.md | #team-engineering, #backend | (none) |
| 2 | Teams/People/@John Smith.md | #team-engineering, #remote | (none) |
| 3 | Misc/Reading.md | #to-read, #book | (none) |

Ask the user: "all", "1,2", "none", or per-file picks.

Step 3: Apply migration

For each confirmed file:

  1. Read the file body, collect all #tag patterns (respecting word boundaries — #tag not https://foo.com/path#section).
  2. Use obsidian property:set name=tags value='["a","b","c"]' type=list path="<file>" to write the frontmatter list. If tags: already exists, merge (don't overwrite). See frontmatter-mutations.
  3. Optionally remove the body #tag patterns (default: leave them — Obsidian indexes both, body tags don't hurt). Add a --strip-body flag for users who want clean bodies.

Step 4: Verify

Run obsidian property:read name=tags path="<file>" on each migrated file to confirm the write. Spot-check that no tags were lost.

Step 5: Commit

See /commit. Commit message: chore: migrate body #tags to frontmatter tags: list (N files).

Important Notes

  • Idempotent: re-running on already-migrated files is a no-op.
  • Don't strip body tags by default — users may rely on visible tags in the body for in-note navigation. Opt-in via --strip-body.
  • Tag normalization: leave tag names as-is. Don't auto-convert #team-engineeringteam-engineering (the leading # is stripped automatically; nothing else).
  • Nested tags (#team/engineering) are preserved.
  • Tags inside code blocks are skipped — they're literal, not Obsidian tags.
  • After running on People files, People.base per-team views can use file.hasTag("team-X") as an alternative to the file.hasLink(file("Teams/+X.md")) filter (which works but is more verbose).

Signals

GitHub stars
165
Forks
13
Last commit
Jun 2026
Advanced
Catalog kind
skill
Gateway key
migrate-tags
Source
github.com/benoror/obsidianos_work