Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Wardrobe Pro: Catalog Your Whole Closet from Photos

    1

    Drop photos from your phone into an inbox and your agent catalogs every garment in them into Obsidian, with your own taxonomy, duplicate detection, "worn with" links and an integrity audit.

    Free

    Security scanned
    Wardrobe Pro: Catalog Your Whole Closet from Photos

    Wardrobe Pro: Catalog Your Whole Closet from Photos

    Example session with this skill installed

    Catalog the photos in my wardrobe inbox.

    • Read your context and instructions
    • Compiled the wardrobe pro
    • Generated the document

    Summary

    6 photos in the inbox: 4 new items, 3 photos linked to items first seen earlier in the same run, 1 proposal for an item you already owned, 1 exact duplicate skipped. Audit: 84 notes, clean.

    Per photo

    PhotoGarmentsDecision
    IMG_0409.jpg (worn)white canvas sneakersnew → SH-004
    burgundy crew-neck teenew → TP-030
    IMG_0412.jpg (worn)grey wide-leg jeansnew → PT-021
    white canvas sneakerssame pair as IMG_0409.jpg → linked to SH-004
    thin silver ringproposal: looks like AC-001 (same thin plain band)
    IMG_0413.jpg (worn)burgundy tee, grey jeanslinked to TP-030 and PT-021
    IMG_0415.jpgsame bytes as an archived photoexact_duplicate_skipped, left alone
    IMG_0418.jpg (flat)sage linen shirtnew → SR-006, colour TBD
    IMG_0420.jpg (worn)the same sage shirtlinked to SR-006

    Proposals

    1. Ring in IMG_0412.jpg → AC-001 Thin silver ring? Same width, no stones, worn on the same finger.

    You: yes. → archived the photo and linked it to AC-001 with --confirmed.

    Enriched

    PT-021 Grey wide-leg jeans,

    TP-030 Burgundy crew-neck tee,

    SH-004 White canvas sneakers,

    SR-006 Sage linen shirt: description, photos, "Worn with" and pairings. Example from PT-021:

    
    ## Worn with
    - [[SH-004|White canvas sneakers]] (2 photos)
    - [[TP-030|Burgundy crew-neck tee]] (1 photos)
    
    ## How to pair
    - With [[TP-011|White ribbed tank]] and [[OW-003|Navy blazer]]: the wide leg balances the blazer's structured shoulders.
    - With [[KN-002|Cream cable-knit sweater]]: soft greys and creams for a tonal autumn look.
    

    Taxonomy

    Queued sage under ## New terms in Taxonomy.md. SR-006 has color: TBD until you move sage into Colors (or pick green).

    Audit

    notes_audited=84, audit=clean.

    Next

    Review the queued term sage, and open Wardrobe.base to see the new items in the table.

    IMG_0409.jpg

    PDF · document

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Wardrobe Pro is the full pipeline behind a real closet catalog. You take photos (garments alone, or mirror selfies wearing several) and they land in Photos/Inbox/, synced from your phone if you link the folder to Google Drive, iCloud or Dropbox. Say "catalog my inbox" and the agent works through every photo.

    What it does

    • Every garment in a photo: tops, bottoms, outerwear, shoes, bags, jewellery. A worn photo of five items links to five notes.
    • Matches across the run: the same jeans in ten selfies become one note with ten photos, not ten notes.
    • Existing items are proposals: when a garment looks like one you already own, the agent asks before linking. Nothing is merged behind your back.
    • Exact duplicates skipped by SHA-256, so re-uploading a photo never creates a second item.
    • Your own taxonomy: categories, subcategories, colours, patterns, styles, seasons, occasions and statuses live in one Taxonomy.md you edit. Values outside it are refused; missing terms are queued for your review.
    • A Python helper does every write: it allocates IDs, writes notes, moves photos and refuses links to items that don't exist. The agent decides; the script keeps the vault consistent, and it won't overwrite a note you are editing in Obsidian at that moment.
    • Rich notes: a description of what is visible, the photos, "Worn with" computed from the photos you actually wore together, and pairing ideas that link only real items.
    • Read-only audit with 27 checks (YAML, IDs, taxonomy values, missing photos, broken links, index integrity). It reports, never repairs; the photo index rebuilds with one command.
    • Obsidian Bases table of the whole closet, grouped by category.

    Reads notes made with Wardrobe Basic, so you keep everything you cataloged before.

    Folder layout

    Wardrobe/
      Items/<ID>.md          one note per garment, e.g. TP-007.md
      Photos/Inbox/          new photos land here
      Photos/Archive/Flat/   garment alone
      Photos/Archive/Worn/   garments on a person
      Taxonomy.md            your controlled vocabularies
      Wardrobe.base          Obsidian Bases view
      System/index.json      SHA-256 of every archived photo
    

    Example prompts

    • "Set up my wardrobe in this vault"
    • "Catalog the photos in my wardrobe inbox"
    • "Audit my wardrobe catalog"
    • "Enrich the note for SK-004 with pairings"
    • "What goes with my burgundy tee?"

    Limitations

    • Needs Python 3.8+ and Pillow (pip install Pillow); HEIC photos also need pillow-heif.
    • Needs an agent that can see images.
    • Brand, size and material are recorded only when visible on a label or stated by you.
    • Built for Obsidian: other Markdown tools work, but the Bases view and photo embeds are Obsidian features.

    Permissions

    Runs a bundled local Python script that reads and writes only inside your wardrobe folder, plus one run-state file in the system temp directory. No network calls.

    Needs Python 3.8+ with Pillow (HEIC photos also need pillow-heif) and an agent that can see images. Brand, size and material are recorded only when visible on a label or stated by you. The Bases view and photo embeds are Obsidian features.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 17 days ago

    • Free with an Agensi account

    Requires

    Python 3.8+
    Pillow
    Obsidian
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Environment Variables

    File Scopes

    Wardrobe/**

    Shell runs the bundled scripts/wardrobe.py. Environment variables WARDROBE_VAULT and WARDROBE_DIR are optional and only point the helper at the vault and wardrobe folder. No network access.

    Listed17 days ago

    What's inside

    Frequently Asked Questions

    Yes. Wardrobe Basic is the one-garment-at-a-time version; Pro adds the photo inbox, multi-garment photos, the taxonomy, the Python helper and the audit. Both are free.

    Python 3.8+ with Pillow (pip install Pillow), plus pillow-heif for HEIC photos, and an agent that can see images and run shell commands, such as Claude Code or Codex CLI. Obsidian 1.9+ for the Bases view.

    Yes. Make Wardrobe/Photos a link to a synced folder (Google Drive, iCloud, Dropbox) and upload to Photos/Inbox from the phone. The next run catalogs whatever is there.

    Every archived photo's SHA-256 is indexed, so the same file is skipped. A garment that looks like one you already own becomes a proposal: it is linked only after you confirm.

    Only when you confirm. Links to existing items need your yes, and enriching an existing note happens only when you ask. The audit is read-only and never repairs anything.

    No. The helper makes no network calls and works only inside your wardrobe folder. Your agent looks at downscaled previews the same way it reads any local image.

    New in Business & Operations