This image is a 4x4 grid of 16 renders of ONE low-poly 3D game CHARACTER on a dark background. The 16 views are a turntable: view 1 is top-left, views run left to right then down, and each step rotates the character 22.5 degrees. Study all 16 views before answering.

What your answer becomes: every line below is stored on this character in a searchable catalog. An AI level builder — not a human — casts people by text ("the old man with the beard", "a woman in a red dress", "a cop"), so your words are the only way this character will ever be found or placed. Report what the PERSON looks like — the things a stranger would notice first — not a caption, not a mood, and nothing about the rendering or the background.

Reply with exactly these fourteen lines, nothing before or after, no markdown. Write "-" only when you truly cannot tell.

what: 2 to 4 words naming the person by their most distinctive look or apparent role, like old bald man, police officer, young woman, orc warrior
cat: character
role: standalone
conn: none
size: tall
age: one word from: child young adult old — grey or white hair, a grey/white beard, baldness or wrinkles read as old
build: one word from: slim average heavy broad
hair: 1 to 3 words: bald, or the hair's length/shape (short long ponytail bun braid curly) and its colour
face: 1 to 3 words for facial features and headwear from: beard moustache stubble glasses sunglasses mask hat helmet cap hood crown clean
job: ONE word for the occupation or role the outfit suggests, like police, farmer, knight, pirate, chef — or "-" if the clothes suggest none
colors: 1 to 3 dominant clothing colours, comma separated, plain colour words
style: 1 or 2 words from: low-poly cartoon realistic toy medieval modern sci-fi rustic
desc: at most 20 words of concrete visual detail: age impression, hair or baldness and its colour, facial hair, anything worn on the head or face, clothing and its colours, anything held or carried, and the occupation or role the outfit suggests
back: at most 16 words for how this character reads FROM BEHIND, small, in motion — the view a player has of their own body for the whole game. Use the rear views in the grid, not the front ones. Say what the SILHOUETTE and the block of colour read as to someone who cannot see the face: what shape the head and body make, the dominant colours from behind, and anything still recognisable from back there (a hat, a pack, a weapon, hair). If nothing distinguishes it from behind, say so plainly — "plain brown box head, no features" is the most useful answer you can give