Inspiration
Among all the useful skills for bills and documents, there should be one for motivation, for lifting your spirits, for joy.
It's a skill for motivation and lifting your spirits, for people who screenshot what they dream of, especially women: we love to celebrate and see what came true. My gallery holds 36,000 photos and videos; about 8,000 of them are wishes, purchases and things that came true. In winter I saved a Hawaiian beach. By summer my documents held the boarding passes, and my gallery held photos from that very shore. In spring, a dream ring; in autumn, an "order delivered" screenshot. I dreamed of a house with a big backyard, and the purchase agreement now sits in my documents. I rarely find time to look back at what came true.
What it does
Mine finds my wish screenshots and plan notes on my NAS and beautifully shows which of them already came true: the "I want this" picture next to the ticket, certificate, order or photo where it became real. It becomes a book of my wishes that came true.
It finds wish screenshots and plan notes, looks for later proof and shows "wanted → came true" pairs to confirm.
I ask "What came true since January 2026?", edit the wish list it found and confirm the pairs. With my permission it saves the "Came True" book and tags the proof. Next time it checks only what's new.
How it is different
Built-in search and UGREEN's 30 official skills don't link pictures to my dreams; each uses one NAS expert. A downloaded picture of the same beach doesn't count: only my camera's photos taken after the wish, and I have the final say.
How it works on the NAS
Only the NAS keeps my gallery, with camera, date and place, beside my documents. Photos: search, date, place, camera model; AI sees what's in a picture and reads text. Documents: proof, with a "what was read" report. Files: notes, tags. Knowledge: the "Came True" book, a new edition per run.
Four NAS experts: photo-agent (ug-photo: search, meta, vision), document-agent (ug-document: qa, extract), file-agent (tags), knowledge-agent (save-text). First a metadata filter: my camera, dated after the wish. Then AI is asked twice, in different words; a pair counts only if both say "yes". Places match by photo location first; then AI checks the photo itself. Inputs: a period and folders. Permission only to save the page, tags and album. No outside services; AI may run in Uliya's cloud.
Where it stands
The skeleton passes the UGREEN kit 0.7.7 checker with no errors. On three open models: "is this a wish?" 14/14 for two, 12/14 for the third; a document as proof 10/10 for all. Built: everything in the package. Not yet tested on a NAS: the copy filter, the picture questions, finding screenshots, Russian text, capture dates and places.
| Check | Result |
|---|---|
| UGREEN kit 0.7.7 checker on the skill package | 0 errors, 0 warnings |
| "Is this a wish?" on held-out screenshots, three open vision models run locally | 14 of 14 for two models, 12 of 14 for the third |
| A document as proof (delivered order, ticket, booking, certificate) | 10 of 10 for all three models |
| On the UGREEN NAS model | not tested yet |
How it will be tested
Test 1: the first run finds the right "wish → proof" pairs. Test 2: after I add an order and two photos, the second run finds only new ones. Edge case: a downloaded photo of the same beach doesn't count. Pass: 7 of 7 pairs found, 0 of 3 fakes counted.
Limits
A wish note just needs a screenshot; it arrives with the photos. Wishes without proof are skipped. AI reads one picture at a time through Uliya's built-in models, which per UGREEN's FAQ may use Uliya's cloud.
Built With
- markdown
- ug-document
- ug-filemgr
- ug-knowledge
- ug-photo
- ugos
- uliya
