agentleFS
eduardoweberai-cloudeduardoweberai-cloud/claude-skills/skills/folder-organizer/SKILL.md

folder-organizer

Skill0 starsChanged 59 days agoMITView on GitHub

Organiza um diretório bagunçado (default: Downloads) classificando arquivos soltos numa taxonomia de pastas existente ou acordada. Use quando pedirem 'organizar/arrumar a pasta Downloads', 'organize my Downloads/Desktop/folder', 'limpar essa pasta', 'clean up this directory', 'sort these files' ou qualquer pedido de arrumar pasta. Sempre roda questionário e mostra plano completo de moves ANTES de tocar em arquivo; nunca deleta sem confirmação; memoriza taxonomia e preferências.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. Minerva67/folder-organizer★ 2same name
    2. gerivdb/SKILLS★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository10 other files it gives its agents.

The file

Show its rules and text

We split this skill into the rules it gives an agent

13 rules. About 0 are common, 11 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈11

    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. Distinctive:

    Ask before acting: Run the questionnaire at the start of every session.

    line 14
  2. Distinctive:

    Never scan-and-move in one shot.

    line 15
  3. Distinctive:

    Plan, then GO: Always present the complete move plan (each file -> destination) and wait for explicit approval before moving anything.

    line 16
  4. Distinctive:

    Never delete without confirmation: Default is to move, not remove.

    line 20
  5. Distinctive:

    If a profile exists, confirm it in one line and only re-ask what changes.

    line 29
  6. Distinctive:

    If not, run the first-run questionnaire.

    line 30
  7. Distinctive:

    Run the scanner (Windows-safe UTF-8.

    line 37
  8. Distinctive:

    Read the JSON; do not re-walk the tree by hand.

    line 46
  9. Distinctive:

    Wait for the user's GO. For ambiguous files, collect decisions in one batch.

    line 57
  10. Very specific:

    Write the approved plan to a JSON file ([{"src": "...", "dest": "..."}, ...].

    line 62
  11. Distinctive:

    Update/create the profile JSON with the confirmed taxonomy, any new rules learned this run, and the chosen policies.

    line 77
  12. Very specific:

    Default target when the user names none: the user's Downloads folder (~/Downloads).

    line 83
  13. Distinctive:

    When stating any path back to the user, give the full absolute path.

    line 85

≈ “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

Folder Organizer

Tidy a messy directory by moving loose files into a clear folder taxonomy. Built to be conservative and consultative: ask first, plan visibly, move only on approval, and make every action reversible.

Operating principles (non-negotiable)

  1. Ask before acting. Run the questionnaire at the start of every session. Never scan-and-move in one shot.
  2. Plan, then GO. Always present the complete move plan (each file -> destination) and wait for explicit approval before moving anything.
  3. Never guess silently. Files that cannot be classified with confidence go to an ambiguous list and are handled per the user's chosen policy.
  4. Never delete without confirmation. Default is to move, not remove. Duplicates are shown first; deletion needs an explicit yes.
  5. Reversible. Moves are executed via the script, which writes an undo file.

Workflow

Phase 0 — Load profile & questionnaire

  • Compute the profile slug for the target folder and look for ~/.claude/folder-organizer/profiles/<slug>.json.
  • If a profile exists, confirm it in one line and only re-ask what changes.
  • If not, run the first-run questionnaire.
  • Question bank, profile schema, and storage path: see references/questionnaire.md. Use the AskUserQuestion tool for the key multiple-choice decisions; keep it to a few questions per message.

Phase 1 — Scan (deterministic)

Run the scanner (Windows-safe UTF-8; handles accented filenames):

PYTHONUTF8=1 python -X utf8 "<skill-dir>/scripts/organize.py" scan "<target>" --json "<tmp>/scan.json"

The scan reports: existing top-level categories and their subfolders (valid destinations), loose files at the root, loose files inside each category, and duplicate candidates (exact = same content; siblings = format/version pairs). Read the JSON; do not re-walk the tree by hand.

Phase 2 — Classify & build the plan

  • Map each loose file to a destination using the saved rules first, then references/classification-heuristics.md.
  • Split into: confident moves and ambiguous (handled per policy).
  • Propose at most one new subfolder per clear project cluster that lacks a home.
  • Present the plan grouped by destination, plus: the ambiguous list (with a guess each), the duplicate/sibling findings, and anything intentionally left in place.

Phase 3 — Approve

  • Wait for the user's GO. For ambiguous files, collect decisions in one batch.
  • Resolve duplicates only after explicit confirmation (default: keep format pairs together, never delete).

Phase 4 — Apply (reversible)

Write the approved plan to a JSON file ([{"src": "...", "dest": "..."}, ...]; dest is the destination FOLDER, not the file path; src/dest may be relative to the target) and run:

PYTHONUTF8=1 python -X utf8 "<skill-dir>/scripts/organize.py" apply "<target>" --plan "<tmp>/plan.json"

This creates destination dirs, moves with no-overwrite, never deletes, and writes .folder-organizer-undo-<timestamp>.json in the target. To reverse:

PYTHONUTF8=1 python -X utf8 "<skill-dir>/scripts/organize.py" undo "<target>/.folder-organizer-undo-<timestamp>.json"

Phase 5 — Save profile & report

  • Update/create the profile JSON with the confirmed taxonomy, any new rules learned this run, and the chosen policies.
  • Report: counts moved, new folders created, files left in place, duplicates found, and the undo-file path.

Notes

  • Default target when the user names none: the user's Downloads folder (~/Downloads).
  • Respect never_touch entries in the profile (e.g. large media folders).
  • When stating any path back to the user, give the full absolute path.
  • The script never decides classifications — that is the agent's job, gated by user approval.

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.