The puzzle behind the goggles

About Miniondle

Who makes it—and why

Miniondle is designed, coded, scheduled, and maintained by an independent fan developer. It began with a simple question: could a visual character guess feel fair even when the answer is not one of the immediately recognizable favorites?

The answer became Miniondle's two-part format. Visual recognition rewards players who notice silhouettes, expressions, clothing, and tiny image details. The clue grid gives everyone a logical route forward. No account is required; the daily game, past puzzle archive, and spoiler-free sharing are available directly in the browser.

How the game is different

How clues are reviewed

A character description is not always enough to judge the exact artwork on screen. Hair is the clearest example: a character may usually have a known hairstyle, but the crop, angle, hat, or pose in one image can hide it. Miniondle therefore reviews hair against the exact image used in that puzzle instead of filling the clue from memory.

1. The source reference and a fingerprint of the reviewed image are recorded.
2. The visible hair is marked as visible, partly visible, covered, or unclear.
3. A reviewed clue label and confidence level are attached to that specific image.
4. If the image changes, its picture-specific review must be updated before it can enter the schedule.

Base images and alternate images are reviewed separately. The remaining categories are checked for consistent game labels as well, but Hair receives the extra image-level audit because it is especially sensitive to what the artwork actually shows.

Alternate images and the 45-day cooldown

The schedule protects variety at the character level: the same character identity cannot appear again within 45 days. Every identity in the current rotation has separately reviewed alternate artwork, and a returning character does not reuse the picture from its first puzzle. Each alternate keeps its own image-specific clue review.

A character may eventually reuse a later alternate when there is no third reviewed picture, but the original puzzle image remains excluded from its repeats. New additions can also debut with a more distinctive alternate, then switch to their base portrait when they return.

Sources, corrections, and image rights

Every answer record keeps a character-page source, the exact image reference, and a reviewed explanation for each scored trait. After a puzzle ends, the result screen shows concise project-written clue summaries and links to the evidence page, character page, and exact displayed image. The separate Clue Lab publishes the full answer-eligible source ledger without revealing the daily answer.

If a clue looks wrong for the picture shown, include the puzzle date, character, category, and what you see when you email a correction. For a source question, credit request, or image-rights concern, use the same address or the contact page. Specific reports can be checked and corrected much faster than a general note.