agentleFS
KubementatKubementat/ai_agent_skills/skills/obsidian-open-loops-collector/SKILL.md

obsidian-open-loops-collector

Skill2 starsChanged 8 months agoMITView on GitHub

Collects and presents all open loops from the Obsidian vault for GTD-style reflection. Aggregates open tasks from 00-Tasks/, dangling thoughts (notes containing TODO/idea/later/?), and stub notes (< 10 lines). Use when the user asks for an open loops review, GTD sweep, what's unfinished, or wants to clear their mental backlog.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Obsidian CLI: always use obsidian

    line 20
  2. Distinctive:

    Run all three passes, then synthesize into one structured report.

    line 24
  3. Distinctive:

    Run these four searches and deduplicate results (a note may appear in multiple):

    line 47
  4. Very specific:

    Use the collect_stubs.sh script.

    line 68
  5. Distinctive:

    After presenting the report, offer:

    line 131

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Open Loops Collector

Performs a full GTD-style sweep of the vault and presents a consolidated view of everything unfinished, dangling, or half-baked.

Vault Layout (this user's specific setup)

  • Tasks: 00-Tasks/ folder — structured task files per domain
    • 01 - Tasks - Inbox.md — general inbox
    • 06 - Tasks - Haus.md — house/home tasks
    • 07 - Tasks - Music.md — music production tasks
    • 10 - Tasks - Beruflich.md — professional tasks
    • TODOs - AKTUELL.md — current high-priority TODOs
    • Work/HUK/ — work project tasks (HUK)
  • Vault path: $HOME/main_vault
  • Obsidian CLI: always use obsidian

Workflow: Three-Pass Collection

Run all three passes, then synthesize into one structured report.

Pass 1 — Open Tasks

# Get total count
obsidian tasks todo total=true

# Get all open tasks grouped by file
obsidian tasks todo verbose=true format=json

Parse JSON output. Group tasks by domain:

  • Inbox / Urgent: files in 00-Tasks/
  • Work: files in Work/
  • Music / Creative: files in 03-denkfabrik/ or 07 - Tasks - Music
  • House / Personal: 06 - Tasks - Haus, Privat/, etc.
  • Projects: everything else

Show top items per group (max 5-7 per group). Skip recurring tasks (🔁) in the main list — note them separately.

Pass 2 — Dangling Thoughts (keyword search)

Run these four searches and deduplicate results (a note may appear in multiple):

obsidian search query="TODO" format=json limit=30
obsidian search query="idea" format=json limit=30
obsidian search query="later" format=json limit=20
obsidian search query="?" format=json limit=20

Filter out:

  • Files already in 00-Tasks/ (covered in Pass 1)
  • Files in copilot-conversations/, ZZ-Archiv/, Evernote Import/
  • Lyrics files (03-denkfabrik/Lyrics/)

Present as a deduplicated list of notes with dangling content. For each, read a snippet to show what is dangling:

obsidian search:context query="TODO" limit=5 format=json

Pass 3 — Stub Notes (incomplete / abandoned)

Use the collect_stubs.sh script.

bash scripts/collect_stubs.sh $HOME/main_vault

This returns lines like: 3\tSome Note.md

Group stubs by folder/domain. Highlight any stubs that look like project or idea notes (not just reference stubs). Read the content of interesting-looking ones:

obsidian read file="Note Name"

Output Format

Present the final report as a structured markdown summary:

## 🔁 Open Loops Report — [today's date]

### 📋 Open Tasks ([total count])

**Inbox & Urgent** ([n] tasks)
- [ ] task text — *[file]*
...

**Work** ([n] tasks)
- ...

**Music & Creative** ([n] tasks)
- ...

**House & Personal** ([n] tasks)
- ...

**Other Projects** ([n] tasks)
- ...

---

### 💭 Dangling Thoughts ([n notes])

Notes containing open questions or ideas not yet captured as tasks:
- **Note Name** — "the dangling snippet..."
- ...

---

### 🗒️ Stub Notes ([n stubs])

Incomplete notes that may need attention or archiving:
- **Note Name** (n lines) — consider: develop or archive?
- ...

---

### 🎯 Suggested Next Actions

Based on the above, suggest 3-5 concrete next actions the user could take to close the most impactful loops.

Interaction Patterns

After presenting the report, offer:

  1. "Dive in" — read any specific note in full
  2. "Close it" — mark tasks as done, move note to archive
  3. "Capture it" — turn a dangling thought into a proper task
  4. "Save report" — write the report as a note in the vault

To save the report:

obsidian create name="Open Loops Report [DATE]" content="[report content]"

Performance Notes

  • Total open tasks is typically ~328 — fetch all at once with format=json, parse client-side
  • Stub script runs locally and is fast (~2s)
  • Dangling thought searches: run all 4 in sequence, deduplicate by filename
  • Skip format=json parsing errors by extracting with regex: \[.*\] (DOTALL)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.